ans_0_1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 日志\n",
    "#### 2017.08.05\n",
    "- 接下来,怎么把o dtype 转换成 uint64 dtype? A: 在插入in和out的长宽时发现应该是直接可以自动识别的.\n",
    "* 建立特征, 把每个link的in 和out 对应长宽也插入columns A: 完成\n",
    "* 把in和out对应的travel_time也最好加进来,但是要看好t0起码接近...很可能不行.\n",
    "* 研究一下t0的规律(下图是在别的notebook里研究的全部数据的t0的规律)\n",
    "![](t0_dist_intvl.png)\n",
    "- 下图是每个link_ID包含时间点数据的数量分布![](t0_dist.png) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "* 看一下travel_time和length,width,t0的关系 \n",
    "- 怎么把datetime64数据变成特征? 做成timedelta然后成为标记在年,月,周,日中位置的数据? 待讨论请教.\n",
    "- 网上找的对时间数据的特征化:离散化，离散化成0-1向量(比如周级别为7维向量);加入是是否是特殊日期的向量(一维，0 or 1), 所以我先把时间数据离散成年,月,日,周,时,分. 或者是utc时间(适用聚类). 或者还可以分桶. \n",
    "- 现有的特征可以套进去GBDT试一下, 但是没有把时间数据作为特征的手段的话, 也不能预测啊\n",
    "\n",
    "#### 2017.08.06\n",
    "- 继续把datetime的dir看完"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 代码实现"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 导入数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "s = pd.read_csv('gy_contest_link_traveltime_training_data.txt',nrows=1000,sep=';')\n",
    "s.to_csv('gy_contest_link_traveltime_training_sample.csv',sep=';')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>link_ID</th>\n",
       "      <th>date</th>\n",
       "      <th>time_interval</th>\n",
       "      <th>travel_time</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9377906285566510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 23:20:00,2016-05-21 23:22:00)</td>\n",
       "      <td>17.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3377906288228510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 18:46:00,2016-05-21 18:48:00)</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3377906284395510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 07:06:00,2016-05-21 07:08:00)</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4377906284959500514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 14:34:00,2016-05-21 14:36:00)</td>\n",
       "      <td>3.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9377906282776510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 05:04:00,2016-05-21 05:06:00)</td>\n",
       "      <td>1.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               link_ID        date                              time_interval  \\\n",
       "0  9377906285566510514  2016-05-21  [2016-05-21 23:20:00,2016-05-21 23:22:00)   \n",
       "1  3377906288228510514  2016-05-21  [2016-05-21 18:46:00,2016-05-21 18:48:00)   \n",
       "2  3377906284395510514  2016-05-21  [2016-05-21 07:06:00,2016-05-21 07:08:00)   \n",
       "3  4377906284959500514  2016-05-21  [2016-05-21 14:34:00,2016-05-21 14:36:00)   \n",
       "4  9377906282776510514  2016-05-21  [2016-05-21 05:04:00,2016-05-21 05:06:00)   \n",
       "\n",
       "   travel_time  \n",
       "0         17.6  \n",
       "1          3.5  \n",
       "2         10.0  \n",
       "3          3.5  \n",
       "4          1.5  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "s = pd.read_csv('gy_contest_link_traveltime_training_sample.csv',sep=';',index_col=0)\n",
    "s.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sample = pd.read_csv('gy_contest_link_traveltime_training_data.txt',nrows=1000,sep=';') # 车流量数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "link_info = pd.read_csv('gy_contest_link_info.txt',sep=';') # 路段id,length,width数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "link_top = pd.read_csv('gy_contest_link_top.txt',sep=';')# 路段拓扑关系数据"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 对sample进行预处理"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sample['id'] = round(sample['link_ID']//(10**6)%(10**4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "tm_itvl=sample['time_interval'].str.replace('[','').str.replace(')','').str.split(',')\n",
    "sample['t0'] = tm_itvl.apply(lambda x: x[0])\n",
    "sample['t1'] = tm_itvl.apply(lambda x: x[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "sample['date'] = pd.to_datetime(sample['date'])\n",
    "sample['t0'] = pd.to_datetime(sample['t0'])\n",
    "sample['t1'] = pd.to_datetime(sample['t1'])\n",
    "sample['delta_t'] = sample['t1']-sample['t0']\n",
    "sample['delta_t']=sample['delta_t'].apply(lambda x:x.value/(6*10**10))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 1000 entries, 0 to 999\n",
      "Data columns (total 8 columns):\n",
      "link_ID          1000 non-null uint64\n",
      "date             1000 non-null datetime64[ns]\n",
      "time_interval    1000 non-null object\n",
      "travel_time      1000 non-null float64\n",
      "id               1000 non-null uint64\n",
      "t0               1000 non-null datetime64[ns]\n",
      "t1               1000 non-null datetime64[ns]\n",
      "delta_t          1000 non-null float64\n",
      "dtypes: datetime64[ns](3), float64(2), object(1), uint64(2)\n",
      "memory usage: 62.6+ KB\n"
     ]
    }
   ],
   "source": [
    "sample.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "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>link_ID</th>\n",
       "      <th>date</th>\n",
       "      <th>time_interval</th>\n",
       "      <th>travel_time</th>\n",
       "      <th>id</th>\n",
       "      <th>t0</th>\n",
       "      <th>t1</th>\n",
       "      <th>delta_t</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9377906285566510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 23:20:00,2016-05-21 23:22:00)</td>\n",
       "      <td>17.6</td>\n",
       "      <td>5566</td>\n",
       "      <td>2016-05-21 23:20:00</td>\n",
       "      <td>2016-05-21 23:22:00</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3377906288228510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 18:46:00,2016-05-21 18:48:00)</td>\n",
       "      <td>3.5</td>\n",
       "      <td>8228</td>\n",
       "      <td>2016-05-21 18:46:00</td>\n",
       "      <td>2016-05-21 18:48:00</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3377906284395510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 07:06:00,2016-05-21 07:08:00)</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4395</td>\n",
       "      <td>2016-05-21 07:06:00</td>\n",
       "      <td>2016-05-21 07:08:00</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4377906284959500514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 14:34:00,2016-05-21 14:36:00)</td>\n",
       "      <td>3.5</td>\n",
       "      <td>4959</td>\n",
       "      <td>2016-05-21 14:34:00</td>\n",
       "      <td>2016-05-21 14:36:00</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9377906282776510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>[2016-05-21 05:04:00,2016-05-21 05:06:00)</td>\n",
       "      <td>1.5</td>\n",
       "      <td>2776</td>\n",
       "      <td>2016-05-21 05:04:00</td>\n",
       "      <td>2016-05-21 05:06:00</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               link_ID       date                              time_interval  \\\n",
       "0  9377906285566510514 2016-05-21  [2016-05-21 23:20:00,2016-05-21 23:22:00)   \n",
       "1  3377906288228510514 2016-05-21  [2016-05-21 18:46:00,2016-05-21 18:48:00)   \n",
       "2  3377906284395510514 2016-05-21  [2016-05-21 07:06:00,2016-05-21 07:08:00)   \n",
       "3  4377906284959500514 2016-05-21  [2016-05-21 14:34:00,2016-05-21 14:36:00)   \n",
       "4  9377906282776510514 2016-05-21  [2016-05-21 05:04:00,2016-05-21 05:06:00)   \n",
       "\n",
       "   travel_time    id                  t0                  t1  delta_t  \n",
       "0         17.6  5566 2016-05-21 23:20:00 2016-05-21 23:22:00      2.0  \n",
       "1          3.5  8228 2016-05-21 18:46:00 2016-05-21 18:48:00      2.0  \n",
       "2         10.0  4395 2016-05-21 07:06:00 2016-05-21 07:08:00      2.0  \n",
       "3          3.5  4959 2016-05-21 14:34:00 2016-05-21 14:36:00      2.0  \n",
       "4          1.5  2776 2016-05-21 05:04:00 2016-05-21 05:06:00      2.0  "
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sample.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 对link_top进行预处理"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "link_top[['in0','in1','in2','in3']]=link_top['in_links'].str.split('#',expand=True) \n",
    "link_top[['out0','out1','out2','out3']]=link_top['out_links'].str.split('#',expand=True) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 132 entries, 0 to 131\n",
      "Data columns (total 11 columns):\n",
      "link_ID      132 non-null uint64\n",
      "in_links     126 non-null object\n",
      "out_links    125 non-null object\n",
      "in0          126 non-null object\n",
      "in1          28 non-null object\n",
      "in2          9 non-null object\n",
      "in3          4 non-null object\n",
      "out0         125 non-null object\n",
      "out1         29 non-null object\n",
      "out2         9 non-null object\n",
      "out3         4 non-null object\n",
      "dtypes: object(10), uint64(1)\n",
      "memory usage: 11.4+ KB\n"
     ]
    }
   ],
   "source": [
    "link_top.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>link_ID</th>\n",
       "      <th>in_links</th>\n",
       "      <th>out_links</th>\n",
       "      <th>in0</th>\n",
       "      <th>in1</th>\n",
       "      <th>in2</th>\n",
       "      <th>in3</th>\n",
       "      <th>out0</th>\n",
       "      <th>out1</th>\n",
       "      <th>out2</th>\n",
       "      <th>out3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4377906289869500514</td>\n",
       "      <td>4377906285525800514</td>\n",
       "      <td>4377906281969500514</td>\n",
       "      <td>4377906285525800514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906281969500514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4377906284594800514</td>\n",
       "      <td>4377906284514600514</td>\n",
       "      <td>4377906285594800514</td>\n",
       "      <td>4377906284514600514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906285594800514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4377906289425800514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4377906284653600514</td>\n",
       "      <td>NaN</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906284653600514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4377906284525800514</td>\n",
       "      <td>4377906281234600514</td>\n",
       "      <td>4377906280334600514</td>\n",
       "      <td>4377906281234600514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906280334600514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4377906284422600514</td>\n",
       "      <td>3377906289434510514#4377906287959500514</td>\n",
       "      <td>4377906283422600514</td>\n",
       "      <td>3377906289434510514</td>\n",
       "      <td>4377906287959500514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906283422600514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               link_ID                                 in_links  \\\n",
       "0  4377906289869500514                      4377906285525800514   \n",
       "1  4377906284594800514                      4377906284514600514   \n",
       "2  4377906289425800514                                      NaN   \n",
       "3  4377906284525800514                      4377906281234600514   \n",
       "4  4377906284422600514  3377906289434510514#4377906287959500514   \n",
       "\n",
       "             out_links                  in0                  in1   in2   in3  \\\n",
       "0  4377906281969500514  4377906285525800514                 None  None  None   \n",
       "1  4377906285594800514  4377906284514600514                 None  None  None   \n",
       "2  4377906284653600514                  NaN                 None  None  None   \n",
       "3  4377906280334600514  4377906281234600514                 None  None  None   \n",
       "4  4377906283422600514  3377906289434510514  4377906287959500514  None  None   \n",
       "\n",
       "                  out0  out1  out2  out3  \n",
       "0  4377906281969500514  None  None  None  \n",
       "1  4377906285594800514  None  None  None  \n",
       "2  4377906284653600514  None  None  None  \n",
       "3  4377906280334600514  None  None  None  \n",
       "4  4377906283422600514  None  None  None  "
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "link_top.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 对link_info进行预处理"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 132 entries, 0 to 131\n",
      "Data columns (total 4 columns):\n",
      "link_ID       132 non-null uint64\n",
      "length        132 non-null int64\n",
      "width         132 non-null int64\n",
      "link_class    132 non-null int64\n",
      "dtypes: int64(3), uint64(1)\n",
      "memory usage: 4.2 KB\n"
     ]
    }
   ],
   "source": [
    "link_info.info()"
   ]
  },
  {
   "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>link_ID</th>\n",
       "      <th>length</th>\n",
       "      <th>width</th>\n",
       "      <th>link_class</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4377906289869500514</td>\n",
       "      <td>57</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>4377906284594800514</td>\n",
       "      <td>247</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4377906289425800514</td>\n",
       "      <td>194</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4377906284525800514</td>\n",
       "      <td>839</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4377906284422600514</td>\n",
       "      <td>55</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               link_ID  length  width  link_class\n",
       "0  4377906289869500514      57      3           1\n",
       "1  4377906284594800514     247      9           1\n",
       "2  4377906289425800514     194      3           1\n",
       "3  4377906284525800514     839      3           1\n",
       "4  4377906284422600514      55     12           1"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "link_info.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 对数据进行合并"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata= sample[['link_ID','date','t0','t1','delta_t','travel_time']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata = pd.merge(alldata, link_info[['link_ID','length','width','link_class']], how='left', on='link_ID')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata = pd.merge(alldata, link_top[['link_ID','in0','in1','in2','in3','out0','out1','out2','out3']], how='left',on='link_ID')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1000 entries, 0 to 999\n",
      "Data columns (total 17 columns):\n",
      "link_ID        1000 non-null uint64\n",
      "date           1000 non-null datetime64[ns]\n",
      "t0             1000 non-null datetime64[ns]\n",
      "t1             1000 non-null datetime64[ns]\n",
      "delta_t        1000 non-null float64\n",
      "travel_time    1000 non-null float64\n",
      "length         1000 non-null int64\n",
      "width          1000 non-null int64\n",
      "link_class     1000 non-null int64\n",
      "in0            962 non-null object\n",
      "in1            238 non-null object\n",
      "in2            79 non-null object\n",
      "in3            33 non-null object\n",
      "out0           929 non-null object\n",
      "out1           222 non-null object\n",
      "out2           66 non-null object\n",
      "out3           32 non-null object\n",
      "dtypes: datetime64[ns](3), float64(2), int64(3), object(8), uint64(1)\n",
      "memory usage: 140.6+ KB\n"
     ]
    }
   ],
   "source": [
    "alldata.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": false
   },
   "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>link_ID</th>\n",
       "      <th>date</th>\n",
       "      <th>t0</th>\n",
       "      <th>t1</th>\n",
       "      <th>delta_t</th>\n",
       "      <th>travel_time</th>\n",
       "      <th>length</th>\n",
       "      <th>width</th>\n",
       "      <th>link_class</th>\n",
       "      <th>in0</th>\n",
       "      <th>in1</th>\n",
       "      <th>in2</th>\n",
       "      <th>in3</th>\n",
       "      <th>out0</th>\n",
       "      <th>out1</th>\n",
       "      <th>out2</th>\n",
       "      <th>out3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9377906285566510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 23:20:00</td>\n",
       "      <td>2016-05-21 23:22:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.6</td>\n",
       "      <td>143</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>4377906283759500514</td>\n",
       "      <td>9377906286566510514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906282532600514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3377906288228510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 18:46:00</td>\n",
       "      <td>2016-05-21 18:48:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>26</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>3377906289228510514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906289663800514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3377906284395510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 07:06:00</td>\n",
       "      <td>2016-05-21 07:08:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>43</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>3377906286918510514</td>\n",
       "      <td>3377906287674510514</td>\n",
       "      <td>4377906287243600514</td>\n",
       "      <td>3377906285934510514</td>\n",
       "      <td>4377906280763800514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4377906284959500514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 14:34:00</td>\n",
       "      <td>2016-05-21 14:36:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>22</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>4377906282969500514</td>\n",
       "      <td>4377906285141600514</td>\n",
       "      <td>4377906289141600514</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906288959500514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9377906282776510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 05:04:00</td>\n",
       "      <td>2016-05-21 05:06:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>9377906283776510514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>4377906288194800514</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               link_ID       date                  t0                  t1  \\\n",
       "0  9377906285566510514 2016-05-21 2016-05-21 23:20:00 2016-05-21 23:22:00   \n",
       "1  3377906288228510514 2016-05-21 2016-05-21 18:46:00 2016-05-21 18:48:00   \n",
       "2  3377906284395510514 2016-05-21 2016-05-21 07:06:00 2016-05-21 07:08:00   \n",
       "3  4377906284959500514 2016-05-21 2016-05-21 14:34:00 2016-05-21 14:36:00   \n",
       "4  9377906282776510514 2016-05-21 2016-05-21 05:04:00 2016-05-21 05:06:00   \n",
       "\n",
       "   delta_t  travel_time  length  width  link_class                  in0  \\\n",
       "0      2.0         17.6     143      9           1  4377906283759500514   \n",
       "1      2.0          3.5      26     12           1  3377906289228510514   \n",
       "2      2.0         10.0      43     12           1  3377906286918510514   \n",
       "3      2.0          3.5      22      9           1  4377906282969500514   \n",
       "4      2.0          1.5       9      9           1  9377906283776510514   \n",
       "\n",
       "                   in1                  in2                  in3  \\\n",
       "0  9377906286566510514                 None                 None   \n",
       "1                 None                 None                 None   \n",
       "2  3377906287674510514  4377906287243600514  3377906285934510514   \n",
       "3  4377906285141600514  4377906289141600514                 None   \n",
       "4                 None                 None                 None   \n",
       "\n",
       "                  out0  out1  out2  out3  \n",
       "0  4377906282532600514  None  None  None  \n",
       "1  4377906289663800514  None  None  None  \n",
       "2  4377906280763800514  None  None  None  \n",
       "3  4377906288959500514  None  None  None  \n",
       "4  4377906288194800514  None  None  None  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alldata.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2016-05-21 07:16:00    6\n",
       "2016-05-21 19:00:00    5\n",
       "2016-05-21 03:58:00    5\n",
       "2016-05-21 16:42:00    5\n",
       "2016-05-21 15:24:00    5\n",
       "2016-05-21 03:56:00    5\n",
       "2016-05-21 23:08:00    5\n",
       "2016-05-21 18:22:00    5\n",
       "2016-05-21 10:48:00    5\n",
       "2016-05-21 14:40:00    4\n",
       "2016-05-21 22:52:00    4\n",
       "2016-05-21 15:38:00    4\n",
       "2016-05-21 09:30:00    4\n",
       "2016-05-21 16:50:00    4\n",
       "2016-05-21 11:58:00    4\n",
       "2016-05-21 19:26:00    4\n",
       "2016-05-21 02:40:00    4\n",
       "2016-05-21 19:30:00    4\n",
       "2016-05-21 00:02:00    4\n",
       "2016-05-21 00:48:00    4\n",
       "2016-05-21 10:06:00    4\n",
       "2016-05-21 15:00:00    4\n",
       "2016-05-21 17:30:00    4\n",
       "2016-05-21 10:16:00    4\n",
       "2016-05-21 10:18:00    4\n",
       "2016-05-21 22:32:00    4\n",
       "2016-05-21 12:46:00    4\n",
       "2016-05-21 17:40:00    4\n",
       "2016-05-21 13:12:00    4\n",
       "2016-05-21 17:46:00    4\n",
       "                      ..\n",
       "2016-05-21 07:04:00    1\n",
       "2016-05-21 13:08:00    1\n",
       "2016-05-21 10:42:00    1\n",
       "2016-05-21 02:10:00    1\n",
       "2016-05-21 00:56:00    1\n",
       "2016-05-21 21:42:00    1\n",
       "2016-05-21 08:16:00    1\n",
       "2016-05-21 16:48:00    1\n",
       "2016-05-21 09:28:00    1\n",
       "2016-05-21 15:36:00    1\n",
       "2016-05-21 19:52:00    1\n",
       "2016-05-21 07:00:00    1\n",
       "2016-05-21 16:46:00    1\n",
       "2016-05-21 04:32:00    1\n",
       "2016-05-21 15:02:00    1\n",
       "2016-05-21 06:58:00    1\n",
       "2016-05-21 16:44:00    1\n",
       "2016-05-21 07:40:00    1\n",
       "2016-05-21 02:04:00    1\n",
       "2016-05-21 19:10:00    1\n",
       "2016-05-21 11:50:00    1\n",
       "2016-05-21 05:50:00    1\n",
       "2016-05-21 19:54:00    1\n",
       "2016-05-21 19:08:00    1\n",
       "2016-05-21 11:48:00    1\n",
       "2016-05-21 05:40:00    1\n",
       "2016-05-21 16:40:00    1\n",
       "2016-05-21 17:28:00    1\n",
       "2016-05-21 07:42:00    1\n",
       "2016-05-21 21:44:00    1\n",
       "Name: t0, Length: 536, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alldata['t0'].value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "## 把每个link的in 和out 对应长宽也插入columns,"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def in_length(x):\n",
    "    try:\n",
    "        return link_info.query(\"link_ID=='%s'\" % x)['length'].iloc[0]\n",
    "    except:\n",
    "        return None\n",
    "alldata['in0_len']=alldata['in0'].apply(in_length)\n",
    "alldata['in1_len']=alldata['in1'].apply(in_length)\n",
    "alldata['in2_len']=alldata['in2'].apply(in_length)\n",
    "alldata['in3_len']=alldata['in3'].apply(in_length)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata['out0_len']=alldata['out0'].apply(in_length)\n",
    "alldata['out1_len']=alldata['out1'].apply(in_length)\n",
    "alldata['out2_len']=alldata['out2'].apply(in_length)\n",
    "alldata['out3_len']=alldata['out3'].apply(in_length)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def in_width(x):\n",
    "    try:\n",
    "        return link_info.query(\"link_ID=='%s'\" % x)['width'].iloc[0]\n",
    "    except:\n",
    "        return None\n",
    "alldata['in0_wid']=alldata['in0'].apply(in_width)\n",
    "alldata['in1_wid']=alldata['in1'].apply(in_width)\n",
    "alldata['in2_wid']=alldata['in2'].apply(in_width)\n",
    "alldata['in3_wid']=alldata['in3'].apply(in_width)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata['out0_wid']=alldata['out0'].apply(in_width)\n",
    "alldata['out1_wid']=alldata['out1'].apply(in_width)\n",
    "alldata['out2_wid']=alldata['out2'].apply(in_width)\n",
    "alldata['out3_wid']=alldata['out3'].apply(in_width)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 研究一下t0的规律"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "% matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x20ba81ade10>]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYwAAAD3CAYAAAAOq2P8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nxBOGc865RP4FKug7OdYofCYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x20ba811ce80>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.plot(alldata['t0'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 每个的link_ID的t0的情况(有多少条)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>t0</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>link_ID</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3377906280028510514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906280395510514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906281518510514</th>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906281774510514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906282328510514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906282418510514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906283328510514</th>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906284028510514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906284395510514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906285434510514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906285934510514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906286918510514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906287674510514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906287886510514</th>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906287934510514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906288228510514</th>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906289044510514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906289228510514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906289434510514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3377906289674510514</th>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280141600514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280163800514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280234600514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280241600514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280329500514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280334600514</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280344800514</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280763800514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280784800514</th>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906280815800514</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288234600514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288243600514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288421600514</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288425800514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288593600514</th>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288663800514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288738500514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288869500514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288893600514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906288959500514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289041600514</th>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289063800514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289141600514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289243600514</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289244800514</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289425800514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289525800514</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289663800514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289813600514</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289863800514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4377906289869500514</th>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906281555510514</th>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906282776510514</th>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906283125510514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906283776510514</th>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906284555510514</th>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906285566510514</th>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906286566510514</th>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906288175510514</th>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9377906289175510514</th>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>130 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                     t0\n",
       "link_ID                \n",
       "3377906280028510514   9\n",
       "3377906280395510514   4\n",
       "3377906281518510514  15\n",
       "3377906281774510514   5\n",
       "3377906282328510514   9\n",
       "3377906282418510514  11\n",
       "3377906283328510514  13\n",
       "3377906284028510514   5\n",
       "3377906284395510514  11\n",
       "3377906285434510514   9\n",
       "3377906285934510514   9\n",
       "3377906286918510514   5\n",
       "3377906287674510514   5\n",
       "3377906287886510514  13\n",
       "3377906287934510514   4\n",
       "3377906288228510514   3\n",
       "3377906289044510514   9\n",
       "3377906289228510514   6\n",
       "3377906289434510514   6\n",
       "3377906289674510514  12\n",
       "4377906280141600514  11\n",
       "4377906280163800514  11\n",
       "4377906280234600514   4\n",
       "4377906280241600514   6\n",
       "4377906280329500514   4\n",
       "4377906280334600514   8\n",
       "4377906280344800514  10\n",
       "4377906280763800514   6\n",
       "4377906280784800514  12\n",
       "4377906280815800514   7\n",
       "...                  ..\n",
       "4377906288234600514   4\n",
       "4377906288243600514   5\n",
       "4377906288421600514  10\n",
       "4377906288425800514   9\n",
       "4377906288593600514   2\n",
       "4377906288663800514   4\n",
       "4377906288738500514   6\n",
       "4377906288869500514   9\n",
       "4377906288893600514  11\n",
       "4377906288959500514  11\n",
       "4377906289041600514  13\n",
       "4377906289063800514   5\n",
       "4377906289141600514   4\n",
       "4377906289243600514   8\n",
       "4377906289244800514   8\n",
       "4377906289425800514   6\n",
       "4377906289525800514   8\n",
       "4377906289663800514   6\n",
       "4377906289813600514  10\n",
       "4377906289863800514  11\n",
       "4377906289869500514   7\n",
       "9377906281555510514   9\n",
       "9377906282776510514  10\n",
       "9377906283125510514   6\n",
       "9377906283776510514   6\n",
       "9377906284555510514   5\n",
       "9377906285566510514  11\n",
       "9377906286566510514   4\n",
       "9377906288175510514  12\n",
       "9377906289175510514   8\n",
       "\n",
       "[130 rows x 1 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alldata[['link_ID','t0']].groupby('link_ID').count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 看一下travel_time和length,width,t0的关系"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x20ba859b0b8>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAasAAAGoCAYAAAD4hcrDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x20ba859bbe0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(alldata['length'],alldata['travel_time'],)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<seaborn.axisgrid.JointGrid at 0x20ba9c79cc0>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAasAAAGoCAYAAAD4hcrDAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAIABJREFUeJzt3Xt4VNW9//F3MrmRKwECVOWIgiyvXMRbtQqltB579KcW\nqy3taY+2iv5EbaVeqvB4ebRSj1WrHm+o9XL0pxb1tLXaUy3ihWqpFW+VLopWrYASQsg9k2SS3x97\ngiHOzJ6QzOy1J5/X8/iY7O9M+M5kMp/Ze6+9Vl5PTw8iIiIuyw+6ARERET8KKxERcZ7CSkREnKew\nEhER5ymsRETEeQVBN7AzurpiPfX1rUG3AUB1dSku9OJKH+BOL670Ae704kofoF4SqampyAu6B1eF\ncs+qoCASdAvbudKLK32AO7240ge404srfYB6kYEJZViJiMjwEsrDgCLZtPL1DWndbvb0XTPcyeD9\n7uX3aWpuT+u2YXg8Mnxoz0pERJynsBIREecprERExHkKKxERcZ7CSkREnKewEhER5ymsRETEeQor\nERFxnsJKREScp7ASERHnKaxERMR5CisREXGewkpERJynsBIREecprERExHkKKxERcZ7CSkREnKew\nEhER5ymsRETEeQorERFxnsJKREScp7ASERHnKaxERMR5CisREXGewkpERJynsBIREecprERExHkK\nKxERcZ7CSkREnKewEhER5ymsRETEeQorERFxnsJKREScp7ASERHnKaxERMR5CisREXGewkpERJyn\nsBIREecprERExHkKKxERcZ7CSkREnKewEhER5ymsRETEeQVBNyCw8vUNad1u9vRdM9yJiIibFFYi\n4hx9gJP+dBhQREScpz0rGXK/e/l9mprbfW+nT8Uiki7tWYmIiPMUViIi4jyFlYiIOE9hJSIizsvr\n6ekJugcREZGUtGclIiLOU1iJiIjzFFYiIuI8hZWIiDhPYSUiIs5TWImIiPMUViIi4jyFlYiIOE9h\nJSIizlNYiYiI80K5nlVXV6ynvr416DYAqK4uxYVeXOkD3OnFlT7AnV5c6QPUSyI1NRV56d62trYp\n5+bKS/X4Q7lnVVAQCbqF7VzpxZU+wJ1eXOkD3OnFlT5AvcjAhDKsRERkeFFYiYiI8xRWIiLiPIWV\niIg4T2ElIiLOU1iJiIjzFFYiIuK8UF4ULCLBefvtt/j5z6+joCDCwQcfxmmnnZHwds8//xzPPfcs\nl19+9fZtsViMyy77McceewLHHXd0oH1Go+1ceeUS6uvrGTmykgsuWEJ1dXXS+91zz528/PJLRCIF\nnHvu+ey77/60tbVx3XXXsGnTRjo7O/nhDy9g333355lnfsejj/4/Cgoi7LnnZBYtupj8fO0bDIae\nPZEUop0xNte3Eu2MBd2KM6677houv/xqbr31bt55523WrfvbZ25z443Xcccdt9DT071924YNH7Fw\n4emsXfuOE30+8cRy9txzMrfeehcnnHAC9913d9L7Wfs3Xn/9Ne688z4uv/wnXH/9tQA89ND97Lnn\nJG699S4uumgxH374AdFoO8uW3cbNN9/BbbfdQ3NzM3/844tZecy5THtWIgnEurt5ZMV61qyrZWtj\nlFGVxcyYUsMpcyYTSeMT8lNP/YYXX1xJa2sr27Zt47zzzuHAAw9nzZq/cOedtxKJRNhll1258MJL\niUbbWbr0Kpqbm9iypZavfe1kTjzxJBYuPIPq6lE0NjayaNGFXHPNlUQiBXR3d3PZZVcxbtx4br75\nBt5883UAvvzlf+Xkk7/J1VdfTmFhIR9/vIm6ui1ccsnlGLM38+Ydy157TWaXXSZw7rmLtvd64YU/\noLX106mGJk7ckx/96OKEj6ulpZnOzg523XU3AA455PO8+upqpkzZe4fbHXDAVI46aja/+tVj27e1\ntrZy0UVLePDB+3a47cKFZ3DLLXd+Ztvuu0/kgw/eB+CKK37C6NFjttcfe+wRnnvuDzvcZ/HiKxk/\nfnzafb755hvMn/8dAI466ihuuumWpPcrLCzi4IMPIy8vj/HjxxOLdVFfX8/q1a/wpS99mfPPX0hp\naRmLFl1EYWERt99+DyUlJYC3N1lUVJzw+ZT0KaxEEnhkxXqeffWj7d/XNUa3fz9/7pS0fkZbWxs3\n3PBfbNtWz5lnnspDDz3OT396NbfddhfV1aNYtuw2nnrqNxizD3PnfoVZs+awZUstCxeewYknngTA\n3LlHM2vWF3nssUfZZ5/9+L//9zzeeGMNLS3NrFr1Ips2beTOO+8lFotx1lnfY+bMgwEYP/5zXHjh\npfz610/w618/zgUXXMLmzZ/wq1/9D11dO/7ZX3vtjWk/Ly0tLZSWlm3/vrS0lI0bN3zmdl/60ld4\n7bVXd9i2116Jn7f+QdVr//2ncsEFl/D447/kgQd+wQ9+cMH22rx5pzBv3imD6rOlpYXy8nIAysrK\naGlpTnq/oqIiqqpG9tnu3b6hYRtNTU1cf/0tPP30k9xyy40sWXIlo0aNBmD58odpa2vj4IMPTdqr\npEdhJdJPtDPGmnW1CWtr1m1h3qxJFBf6zyU3ffqB5OfnM2rUaCorK9mypZa6ui0sWeLttUSjUQ4+\n+FA+//kjePTRh3j++ecoLS2jq6tr+8/4l3/ZHYBjjz2eBx+8j0WLzqGsrJwFC87mgw/+wbRp08nL\ny6OgoID99juA999/D4C99jIAjB07jrfeegOAqqqRVFdXU1vbtEOffntWffdiFi++gra2T2/b2tpK\neXmF73OxM3qD94ADpvLSS8/vUPPbsyorK/Pts6ysjNbWFuDT4Ep2v8LCwu239ba3UF5eQWVlFUcc\ncRQARxxx1Pa9xu7ubm699Sb++c8PuPrqa8nLS3t+WklCYSXST0NzlK2N0YS1+qZ2GpqjjK0u9f05\n1nrnSLZuraO5uZmamrGMHTuWpUuvp7y8nJdeep4RI0p5+OH/Zv/9p3LiiSfx2muv8vLLL23/Gb0n\n5V966XmmTZvBaaedwTPP/I4HH7yPWbPm8NRTv+aUU75FV1cXb7/9Jscccyzwx4RvjslO8PvtWfXf\niykoKGTDho/YZZddWb36ZU49NfEAi8Gydi1jx47jzTffYI899kzZU39lZeW+fR5wwDRefnkV++67\nPy+88ALTps1Ier9IJMJtt93EN7/572zevJnu7h5GjhzJ1KnTeeWVVey99z688cZrTJzo9fmf//kT\nCgsLueaan2lgxRBRWIn0U1VezKjKYuoSBFZ1RQlV5emdf9i6tY7zzjuL5uZmLrvsMiKRCOed9yMu\nuOA8enp6KC0tY8mSK8jLy+OGG67lD3/4PeXl5UQiETo6Onb4WXvvvS9XXXUZ9913N93d3ZxzzvkY\nszdr1vyFBQtOpbOzkzlz5mLM3km6GTo/+tGPueKKxXR3d3PwwYey3377A/DDH57NtdfeSGFh4YB+\nXqJzVgBPPfUkjzzyECUlJSxZcuWQ93niiSdx1VWXcdZZ36O0tIRLLrki5f2mTp3OggWn0tPTw/nn\nXwTAd75zKkuXXsWCBadSUFDA4sVXYO3fePLJXzFt2gzOPfdMAL7+9W8ya9YXB/wY5FN5PT2hXBKl\np/+hjKDU1FR85rDKcO4D3OllMH089Oy6Hc5Z9Zp70G5pnbN66qnf8MEH73PWWecMupeh5Eof8Gkv\nP//5zzjvvEU71BYuPIMLLriE3XefmNVegqb1rJI/fu1ZiSRwypzJgHeOqr6pneqKEmZMGbN9uwyd\nb3zjW0G3ICGgsBJJIJKfz/y5U5g3axINzVGqyovTGlTR66tfPS6D3eWWcePGf2ZbshGCMnwprERS\nKC6MpDWYQkQyS8NURETEeQorERFxnsJKREScp7ASERHnKaxERMR5CisREXGewkpERJynsBIREedl\n9KJgY8xrQGP8238AVwP3Aj3A28DZ1tpuY8zpwAKgC7jKWvtkJvsSEZFwyVhYGWNKgDxr7ew+234N\nLLbWrjTG3A4cb4x5GTgXOAgoAV4yxjxjrU28RoOIiAw7mdyzmgaUGmN+H/93LgFmAr2rqD0NfAWI\nAavi4RQ1xqwHpgJ/TvXDa2oys+DbznClF1f6AHd6caUPcKcXV/oA9TIY1dWlFBSkP19l2GUyrFqB\n64C7gL3wwinPWts7rX0TUAVUAg197te7PSUXpvMHp5YWcKIPcKcXV/oAd3pxpQ9QL8n6SFd9fav/\njUIm1ePPZFitA9bHw2mdMaYOb8+qVwWwDe+cVkWC7SIiIkBmRwOeBvwMwBizC94e1O+NMbPj9WOA\nF4HVwJHGmBJjTBWwD97gCxERESCze1Z3A/caY17CG/13GrAFWGaMKQLWAsuttTFjzE14wZUPXGqt\nbc9gXyIiEjIZCytrbQcwP0FpVoLbLgOWZaoXEREJN10ULCIizlNYiYiI8xRWIiLiPIWViIg4T2El\nIiLOU1iJiIjzFFYiIuI8hZWIiDhPYSUiIs5TWImIiPMUViIi4jyFlYiIOE9hJSIizlNYiYiI8xRW\nIiLiPIWViIg4T2ElIiLOU1iJiIjzFFYiIuI8hZWIiDhPYSUiIs5TWImIiPMUViIi4jyFlUgK0c4Y\nm+tbiXbGgm5FZFgrCLoBERfFurt5ZMV61qyrZWtjlFGVxcyYUsMpcyYTyddnPJFsU1iJJPDIivU8\n++pH27+va4xu/37+3ClBtSUybOkjokg/0c4Ya9bVJqytWbdFhwRFAqCwEumnoTnK1sZowlp9UzsN\nzYlrIpI5CiuRfqrKixlVWZywVl1RQlV54pqIZI7CSqSf4sIIM6bUJKzNmDKG4sJIljsSEQ2wEEng\nlDmTAe8cVX1TO9UVJcyYMmb7dhHJLoWVSAKR/Hzmz53CvFmTaGiOUlVerD0qkQAprERSKC6MMLa6\nNOg2RIY9nbMSERHnKaxERMR5CisREXGewkpERJynsBIREecprERExHkKKxERcZ7CSkREnKewEhER\n5ymsRETEeQorERFxnsJKREScp7ASERHnZXTWdWPMWOAvwJeBLuBeoAd4GzjbWtttjDkdWBCvX2Wt\nfTKTPYmISPhkbM/KGFMI3AG0xTddDyy21h4J5AHHG2PGA+cCRwBHA9cYY7RmuIiI7CCThwGvA24H\nNsa/nwk8H//6aWAucAiwylobtdY2AOuBqRnsSUREQigjhwGNMf8B1Fpr/9cY8+P45jxrbU/86yag\nCqgEGvrctXe7r5qaiiHqdvBc6cWVPsCdXlzpA9zpxZU+QL0MRnV1KQUFw2f16kydszoN6DHGzAWm\nA/cDY/vUK4BtQGP86/7bfdXWNg1Np4NUU1PhRC+u9AHu9OJKH+BOL670AeolWR/pqq9vzWAnwUj1\n+DMSVtbao3q/NsasBM4E/tMYM9tauxI4BngOWA1cbYwpAYqBffAGX4iIiGyX0dGA/SwClhljioC1\nwHJrbcwYcxPwIt75s0utte1Z7ElEREIg42FlrZ3d59tZCerLgGWZ7kNERMJLFwWLiIjzFFYiIuI8\nhZWIiDhPYSUiIs5TWImIiPMUViIi4jyFlYiIOE9hJSIizlNYiYiI8xRWIiLiPIWViIg4T2ElIiLO\nU1iJiIjzFFYiIuI8hZWIiDhPYSUiIs5TWImIiPMUViIi4jyFlYiIOE9hJSIizlNYiYiI8xRWIiLi\nPIWViIg4T2ElIiLOU1iJiIjzFFYiKUQ7Y2yubyXaGQu6FZFhrSDoBkRcFOvu5pEV61mzrpatjVFG\nVRYzY0oNp8yZTCRfn/FEsk1hJZLAIyvW8+yrH23/vq4xuv37+XOnBNWWyLClj4gi/UQ7Y6xZV5uw\ntmbdFh0SFAmAwkqkn4bmKFsbowlr9U3tNDQnrolI5iisRPqpKi9mVGVxwlp1RQlV5YlrIpI5CiuR\nfooLI8yYUpOwNmPKGIoLI1nuSEQ0wEIkgVPmTAa8c1T1Te1UV5QwY8qY7dtFJLsUViIJRPLzmT93\nCvNmTaKhOUpVebH2qEQCpLASSaG4MMLY6tKg2xAZ9nTOSkREnKewEhER5ymsRETEeQorERFxnsJK\nREScp7ASERHnKaxERMR5CiuRFLT4oogbfC8KNsYUARcABlgI/ABYaq3tyHBvIoHR4osibknnr+6/\ngDLgQKALmAzcncmmRILWu/hiXWOUHj5dfPGRFeuDbk1kWEonrGZaay8BOq21rcB3gRmZbUskOFp8\nUcQ96cwN2BM/FNgT/35Mn6+TMsZEgGV4hw97gDOBduDe+PdvA2dba7uNMacDC/D23K6y1j45wMch\nMmTSWXxR8wWKZFc6e1Y3As8C440xNwKvAjekcb/jAKy1RwCLgauB64HF1tojgTzgeGPMeOBc4Ajg\naOAaY4xWt5PAaPFFEff4hpW19gG8vaKrgfeA46y196Rxv/8Bzoh/uzuwDZgJPB/f9jQwFzgEWGWt\njVprG4D1wNQBPg6RIaPFF0Xck+5owElAU3zTdGPMdGvt/X73tdZ2GWPuA04ETgK+bK3tPYTYBFQB\nlUBDn7v1bk+ppqbC7yZZ40ovrvQB7vSys30sPHkGpSOKeOXtTWzZ1saYkSM4bP/Pcdpx+xGJ7Nxo\nwLA/J5mgXnZedXUpBQXD54NTOuesnsY7ZPdBn209gG9YAVhrv2uMuQj4EzCiT6kCb2+rMf51/+0p\n1dY2+d0kK2pqKpzoxZU+wJ1eBtvHCUdM5JhDJuyw+OLWrS2B9DJUXOkD1EuyPtJVX9+awU6Ckerx\npxNWY6y10wb6jxpj/h3YzVp7DdAKdAOvGmNmW2tXAscAzwGrgauNMSVAMbAP3uALkcBp8UURN6QT\nViuMMXOBFdba7gH87MeBXxhjXgAK8S4mXgssix9aXAsst9bGjDE3AS/inUO71FrbPqBHISIiOS2d\nsPoQ+D3eEHbwDgn2WGtTHiy11rYAJycozUpw22V4w9xFREQ+I52wOg+YaK39MNPNiLgm2hnb4ZyV\niAQjnbDaANRluhERl2huQBG3pBtWbxtjVgHbJ6+11p6Wsa5EAtY7N2Cv3rkBAebPnRJUWyLDVjph\n9dv4fyLDgt/cgPNmTdIhQZEsSxpWxpjx1tqP8YaXiwwbmhtQxD2p9qzuAo7Fmx6pB28UYK8eYM8M\n9iUSmN65AesSBJbmBhQJRtKwstYeG/9yprV2a9+aMWZiJpsSCVLv3IB9z1n10tyAIsFIdRhwAt7e\n1FPGmGP4dM+qAHgK2Dvz7YkE45Q5kwHvHFV9UzvVFSXMmDJm+3YRya5UhwGvAL4I7AK80Gd7F6D1\npiSnRfLzmT93CvNmTdJ1ViIOSHUY8DQAY8xF1tqfJrqNMeZYLZQouUxzA4q4IZ31rBIGVdyVQ9iL\niIhIQoO9FD/P/yYiIiKDM9iw6vG/iYiIyOBokjMREXGewkpERJync1YiIuK8VBcFfyfVHa219wOf\nH/KORERE+kl1UfAXU9R6gPu1/LzkOi2+KOKGVBcFn9r3e2NMtbW2PvMtiQRPiy+KuMV3PStjzDTg\nEaDUGHMY3tRLJ1trX8t0cyJB0eKLIm5J5yPizcCJQJ21diNwFnB7RrsSCZDf4ovRzliWOxKRdMKq\n1Fq7tvcba+0zgBb0kZyVzuKLIpJd6YTV1vihwB4AY8y3gK2p7yISXr2LLyaixRdFgpFOWJ0F/Bew\nnzFmG/ADYEFGuxIJUO/ii4lo8UWRYPgOsAAqrLVfMMaUARFrbWOmmxIJmhZfFHFLOmG1zBhTAjwY\n/09hJTlPiy+KuCWd9awOBr4GFOItcb/SGPO9jHc2TEQ7Y2yub9UIM0f1Lr6ooBIJVjp7Vlhr/26M\nuR54F1gEXAzcncnGcp0uOhURSV86FwV/DfgmcCjwJHCOtfaPmW4s1+miUxGR9KWzZ/Ut4AFgvrW2\nM8P9DAt+F53OmzVJh51ERPrwDStr7TxjzAzgMGNMHhAB9rDW3pPx7nJUOhedjq0uzXJXIiLuSucw\n4H3A4cAoYC0wHVgFKKx2Uu9Fp3UJAksXnYqIfFY6Z/KPAvYFfgmcgXfuqiiTTeU6XXQqQYh2xti0\npUUjTyWU0jlntdFa22mMWQtMtdY+bIypyHRjuU4XnUq27DDytCnKqAqNPJXwSSesNhhjfgw8C1xr\njAEoz2hXw4AuOpVs0chTyQXpfKz6HvAPa+2fgcfxhrGfldGuhhFddCqZpOVOJFeks2f1mLX2KwDW\n2pvx1rcSkRDQyFPJFensWY0wxkzIeCciMuS03InkinT2rMYAHxhjPgHagDyg21o7KaOdicig9Y48\n7XvOqpdGnkqYpBNW7wBH44VUT/z/v8hkUyIydDTyVHJB0rAyxjwBTAN2AWb0u8+HGe5LRIZI35Gn\nkaJCYh2d2qOS0Em1Z/VdvFkrfg6c22d7F/BJJpsSkaFXXBihZkwZtbVNQbciMmBJwyq+InAjcHz2\n2hEREfksXb4uIiLOU1iJiIjzFFYiIuK8tJa13xnGmEK8ZUQmAsXAVXjD4O/FGwL/NnC2tbbbGHM6\nsABv8MZV1tonM9WXyEBEO2Oau1HEARkLK+DbQJ219t+NMaOA1+P/LbbWrjTG3A4cb4x5GW+04UFA\nCfCSMeYZa23iOWJEsmCHmcobo4yq1EzlIkHKZFj9Elge/zoPb69pJvB8fNvTwFeAGLAqHk5RY8x6\nYCrw5wz2JpKSZioXcUvGwspa2wwQX/tqObAYuM5a2xO/SRNQBVQCDX3u2rs9pZoad5bUcqUXV/oA\nd3rZmT7aO7p48926hLU3361jwbwRlBQN/E8nzM9JpqiXnVddXUpBwfA5NJ3JPSviE+A+AdxqrX3I\nGHNtn3IFsA3vWq6KBNtTcuXCxpqaCid6caUPcKeXne1jc30rtfVtCWtbtrXx7vt1A56pPOzPSSao\nl8R9pKu+vjWDnQQj1ePP2MF3Y8w44PfARdbae+Kb1xhjZse/PgZ4EVgNHGmMKTHGVAH74A2+EAmE\nZioXcU8mzxRfAlQDS4wxK40xK/EOBV4RH1RRBCy31n4M3IQXXCuAS6217RnsSySl3pnKE9FM5SLB\nyOQ5q/OA8xKUZiW47TJgWaZ6ERkozVQu4paMnrMSCau+M5XrOiuR4OmCkYBFO2Nsrm8l2hkLuhVJ\noLgwwtjqUgWVSMC0ZxUQXXQqIpI+hVVAdNGpiEj69BE+ANHOGGvW1SasrVm3RYcERUT6UVgFoKE5\nytbGxFMf1je109CsaRFFRPpSWAVAF52KiAyMwioAuuhURGRgNMAiILroVEQkfQqrgOiiUxGR9Cms\nAtZ70amIiCSnc1YiIuI8hZWIiDhPYRUwzQ0oIuJP56wCorkBRUTSp7AKiOYGFBFJnz7CB0BzA4qI\nDIzCKgCaG1BEZGAUVgHQ3IAiIgOjsAqA5gYUERkYDbAISO8cgK/ZWuqbolRXFHOgqdHcgCIiCWjP\nKmB5eTv+X0REPkt7VgHR0HURkfRpzyoAGrouIjIwCqsAaOi6iMjAKKwCoKHrIiIDo7AKgIaui4gM\njAZYBETL2ouIpE9hFRAtay8ikj6FVcC0rL2IiD+dsxIREecprERExHkKKxERcZ7CKmDRzhib61s1\na4VkXLQzxqYtLXqtSShpgEVAYt3dPLJiPWvW1bK1McqoymJmTPFmXY/k6zOEDJ0dXmtNUUZV6LUm\n4aOwCogmsg2HaGcs9JcW6LUmuUBhFQC/iWznzZoU2jfGXJEre756rUmuCM9fXQ7RRLbu690bqWuM\n0sOneyOPrFgfdGsDotea5AqFVQA0ka3bcmkJF73WJFcorAKgiWzd1tAcpS7J3sjWxnDtjRQXRpi2\n15iEtWl7jdZrTUJD56wCools3VVVXkxJUT7tHd2fqRUXRUK3N5I3wO1hkQuDXyR9CquAaCJb14X9\nrdwT7Yzx+t+3JKy9/vc6TpodC93rLlcGv8jA6DcbsN6JbMP2hpHLGpqjRDsSn5fqiH+aD4tcOqTZ\nK1cGv8jAKKxE+smlQQm9hzQTCeMhzVwa/CIDo7AS6Sf3BsDkxiFN0FD84UxhJZLASbP3ZMLYcvLj\n7/P5eTBhbDknzd4z2MYGKJcOaUJu7fXKwGQ0rIwxhxpjVsa/nmyMeckY86Ix5jZjTH58++nGmFeN\nMa8YY47NZD8i6Vq+8j3+ubmZ7h7v++4e+OfmZpavfC/YxgaoqryY6oqihLWR5cWhe3PPvb1eSVfG\nwsoYcyFwF1AS33Q9sNhaeyTecYnjjTHjgXOBI4CjgWuMMeH665GcE+2M8ZrdnLD2mq0N1XmR4sII\npSMKE9ZKRxSE8s39lDmTmXvQboyuLCE/D0ZXljD3oN102UeOy+TQ9XeBrwEPxL+fCTwf//pp4CtA\nDFhlrY0CUWPMemAq8OcM9iWSUkNzlK1NHQlrW5uiNDRHGVtdmuWudk60M0ZtfVvCWm19G9HO8A1d\n12Ufw1PGwspa+5gxZmKfTXnW2vhBFZqAKqASaOhzm97tvmpqKoaizSHhSi8u9NHe0cWmLS1UV42g\npCj4y/h25jkpGlFEfj50f/aaYPLzYbddRu7U4bMgfj/vb2og2pnggQDRzm668vLYLcDXzWCfk92G\nqA9w4+9nIKqrSykoGD4hnc13k75/MRXANqAx/nX/7b5qa5uGrrNBqKmpcKKXoPtwcc2knX1ONte3\nJgwq8ALso43b6BjgnlVQv5+PNqb+c/po4zbKCsL1+8kEV3oZSGDW17dmsJNgpHr82XyVrjHGzI5/\nfQzwIrAaONIYU2KMqQL2Ad7OYk8yRHa4ULMn3BdqVpUXMzrJiLPRleEalJBsryrdurhr5esbgm4h\nq7IZVouAK4wxLwNFwHJr7cfATXjBtQK41FrbnsWeApcLy9rn2oWaxYURDpg8OmHtgMnhmvx1j89V\nDqou4oqMHga01r4PHBb/eh0wK8FtlgHLMtmHi3JpfrNUU/rUxaf0CcuAhF72g60D2u6qosIIeUBP\nglpevC4SBuF6V8whuTS/WVV5MZEkr6RIPqE6bAbQ1NrBprrEO/ib6tppak08UtBFDc3RhEEFXoCF\n7aJgGb4UVgHItcNmHZ0xYklOfcS6vXqY/PW91HtPfnWXRPJTT7XkVxdxhcIqALk2v9lHm5sHVXdN\na7RzUHUkFLJ7AAAPoklEQVSXbE5yjVW6dRFXKKwCkGvzm40sTzydT7p110ybnHhl3XTrLhlbPWJQ\ndRFXKKwCkGvzm21rTn0Ox6/uGr9BB2EalBDrTnbGKr26iCsUVgHJpfnNcu3T+7oP6wdVd0lVeTEj\nyxIP+h1ZVhC6vXgZvoKfD2eYyqX5zXLt03tPXupBB351lxQXRujsSvz8d3b1hPY1J8OP9qwClgvL\n2ntBm2w12vzQfXo3E0YOqu6SptYOWqKJR2O2RGOhGoYvw5vCKmC5MIMFQLKdjbwcWqU2jHLpkKYM\nbzoMGJBcm8GivSPJzN4dsdDNYJHOUPx9Jo7KUjeD0xFLfQjWry7iinC9K+aQXJvBItnEr6NCNvEr\nQGlJ6kOyfnWXjPcZ3OJXF3GFwioAuTaDRXFhhNKSJKvRlhSG7nzcuxsaB1V3yS415UlreT51EZco\nrAKQazNYRDtjtLQlPlHf0tYZuvBNFrzp1l2T7I9cZxMlTBRWAUg9g0X4DpulWga+Pr4MfJiMqUr9\n/PvVXVJb30qyFau643WRMFBYBSDXDptVlRdTUpRs6HokdOH713/4TGTrU3dJa3vXoOoirlBYBSDX\nDpt5cueg0ujK1IMO/Oouae9IHUZ+dZflymUfkh4NXQ9AQ3OU+iSHzbY1R0M31LuhOUq0I/EbRkdn\n+Iaujx/tM4LOp+6SXX0GUPjVXZRLl31I+vSbDUCuzbqea+fg1n+UerSfX90luTYVFsDDf/h7wss+\nHv7D34NuTTJIYRWAXJt1vbgwwoiSxDvpI0oKQvd49pmYejolv7pLRhSnPnjiV3dNtDPGqrc2Jayt\nemuTDgnmMIVVQHJp1vVoZ4zaJIv41da3he4NJBJJHa5+dZfUbks92s+v7prabW1JZ0tp7+imdpsW\nk8xV4fpYlUNyadb12m1tRDuTTLfU6b2B7BaicyPJBr+kW3dJsnOj6dZd09GV+oOPXz3XrHx9A7On\n7xp0G1mhsApY76zrodbjc97Dr+6YXHqDr65IvUqzX901RZHUB4P86hJe+s3KoNVUlya9zqqkKEJN\n2MLYL1tDlL01I1M/935119RUlyaf4T+P8L3WJG0Kq4DlwrUixYURDj/gcwlrhx8wPnSHN6uTjGxM\nt+4Sv9lDwja7CCS/oi93rvSTRHQYMCC5dq3IvFl7svqdT2hu+/Qi0/IRBcybtWeAXe2c1vbOQdVd\n0hFLNtlSenXX1Na3kmy0fXePV99tbEV2m5KsCN+7Yo7IpSVCAJb+95odggqgua2Lpf+9JqCOdt5f\n/1E3qLpLWtt8gten7prOrtTh6leX8FJYBSDXlghpau1gQ23iBQs31DaHbun0ZLNxpFt3yboPtw2q\n7prCgtRvWX71XLTy9Q07/Jerht9v1gG5tkTIR5ubUx6a8Vt51zW5dM5qY13q66j86q7xmw0lbLOl\nSPoUVgHItemWxvqsNutXd82b76Y+zOdXd0qez9BFv7pjNm1pGVRdwkthFYBcm26pLZp65m6/umui\nPjOR+9VdUlma+joqv7prPkpyuDnduoSXwiogJ83ekwljy8mPj7fNz4MJY8s5aXb4Rs8lvfAl3bpj\nSopSrwTsV3dJQ0vq84V+ddeM87mOyq8u4aWwCsjyle/xzz7nerp74J+bm1m+8r1gG9sJNSNHUFKU\neG+wpChCzchwHQbMqclfc2x2kdIkEyanWx8OcnWQhcIqAKlHA9aGbjRgcWGEw/cfl7B2+P7jQndY\nc0Rx6j8Lv7pTcmyvV6MBhy/9ZgPQ0BylLslowLrGaOhGAwLJ3/RC9mYI0Oyz1Ltf3SWtbalfS351\n19RUl1KcJJCKC/I13VIOU1gFIKcOM+HtKf5l7ccJa39Z+0no9hS3Nacezu1Xd8nIytSHYP3qriku\njDBmZEnC2piRJaHbi5f0KawCUNuQes0dv7prGpqjNLQmDqSG1q7Q7Sm2Jnks6dZd0tbmM1LTp+6a\naGeMTUmuDdtUF+45NiU1hVUAtja2D6rumrZo6il7/Oqu8ZuaMUxTN+ZHUh+G9au7ZmNt6gvQN2ro\nes4K1/GmHBGNpp6/zK/umj/9dbNvfffxVVnqZvCafT4r+NVdkmvXwCWb+aVvfY9dstSMwwY6IjAM\nCziG6DNi7vjrP7YMqu6aWHfqQy9+ddfk0HJWtPsMBvGru6a8NPU1bn51CS+FVQA6fGaG9qu75oBJ\niWfjSLcumVO7LfVgEL+6azTr+vClsArAmMrEo5nSrbtm9/Gp1w/yq0vmtPqcLvSru2bdh/WDqkt4\nKawCUDoi9aEKv7prcu28iLir3mdkqV9dwkthFYBop89EqT5119T5DLX3q4ukq2xE6jFhfnUJL/1m\nA/DO+6kPVfjVXfPCGxt96/tMHJ2lbiSXfbw19Tk2v7okFvR8gumMRtSeVQCaW1MfqvCru2ZMlc85\nOJ+6SLpyaRVnGRiFVQBKfKZT8qu7ZvaM3QZVF0mXRgMOX068Kxpj8oFbgWlAFPi+tXZ9sF1lzuiK\nEWzYkvzK0tEV4ZqvbXRV6n796iLpGlleDCSfpWJkyFbZlvS5smd1AlBirf08cDHws4D7yahdx5QN\nqu6aTVtST3HjVxdJ1xEHpJ6ewq8u4eXEnhXwBeB3ANbaV4wxBwXcT0ZNGF85qLpr3lhf51v/3Jjy\nLHUjuWzfPUYNqp5LwjBF0lByJawqgYY+38eMMQXW2qRjuGtq3LnQdKC9HHVQEXf+5p0U9X+haicO\nZwT1nMw+ZHceXfluynpQve3Mv7vfxJH89f1tKes783ODeA4KgVTX/RYS7N/SzvzbXzl0Ar//0z8T\nbt9tl5FZ7SVI1dWlFBQMnyVRXAmrRqDvKyU/VVAB1NY2ZbajNNXUVOxUL7vWlLGhtiXh9o62Dmrb\nOrLSx1AoyYeCSB5dsc/OmlcQyaMkP5jf184+J4u+cSCnLV2Rsj7QnxvU7+eOi+ekfCx3XDwnsNfN\nzj4nX581ie5YD39e+wkNLZ1UlRVy8D7j+PqsSTv9WIL8++nfR7rq63NvmH6qx+/KOatVwFcBjDGH\nAW8F207mLfnuTCaMLSc/vkJDfh5MGFvOku/ODLaxnXT9OUdQ0G+5iYJIHtefc0RAHQ3O17+wx4C2\nu+yA3RN/Jk223XWR/Hzmz53C0jMPZ+mCw1h65uHMnzuFSJjWbpEBy+vpCX4O6T6jAacCecCp1tq/\npbhLjwufgmDwn8iaWjv4aHMzu40tp6K0KLA+hsqmLc2s/7iFyePLAj9PNRTPydL7VrNuUzNTPlfO\nxd89JNBeBuv7S1fQjfcJ9a6L5wTaC7jxnPRypZeamoq0FxirrW0K/s17iKV6/E6E1U7ImbDKtT7A\nnV5c6QPc6cWVPkC9JOlDYZWE9ptFRMR5CisREXGewkpERJynsBIREecprERExHkKKxERcZ7CSkRE\nnKewEhER5ymsRETEeWGdwUJERIYR7VmJiIjzFFYiIuI8hZWIiDhPYSUiIs5TWImIiPMUViIi4jyF\nlYiIOK8g6AYGwhiTD9wKTAOiwPettesD6KMQuAeYCBQDV1lrf53tPvr1NBb4C/Bla+3fAurhx8D/\nAYqAW621dwfURyFwH97vJwacnu3nxBhzKPBTa+1sY8xk4F6gB3gbONta2x1QL9OBm/GelyjwHWvt\nJ0H00mfbfOAca+3ng+gj/rezDKgGInjPybsB9TIduB3oAtbhvcdl7bXisrDtWZ0AlMRf1BcDPwuo\nj28DddbaI4F/BW4JqA9g+5vzHUBbgD3MBg4HjgBmAROC6gX4KlBgrT0cuBK4Opv/uDHmQuAuoCS+\n6Xpgcfz1kgccH2AvP8cLhtnA48BFAfaCMWYG8D285yWoPq4FHrTWHgUsBvYOsJfLgCuttV/A+yD8\nb9nqxXVhC6svAL8DsNa+AhwUUB+/BJbEv87D+xQUpOvwPo1tDLCHo4G3gCeA3wBPBtjLOqAgvide\nCXRm+d9/F/han+9nAs/Hv34amBtgL9+w1r4e/7oAaA+qF2PMaOAnwA+y2MNn+sD7gLWbMeZZ4FvA\nygB7WQOMMsbkARVk/7XrrLCFVSXQ0Of7mDEm64cyrbXN1tomY0wFsBzv01ggjDH/AdRaa/83qB7i\nxuB9ePg6cCbwYPwPLgjNeIcA/4Z3eOembP7j1trH2PFNJs9a2zuvWRNQFVQv1tpNAMaYw4GFwA1B\n9GKMiQB3A+fjPSdZk+D3MxGot9bOBT4ki3ubCXr5O97rdS0wjuwGp9PCFlaNeJ82euVbawPZqzHG\nTACeAx6w1j4URA9xpwFfNsasBKYD9xtjxgfQRx3wv9baDmutxfvEXhNAHwA/jPcyBe/85n3GmBKf\n+2RS33MOFcC2oBoBMMacgrcn/m/W2tqA2pgJ7AXcBjwM7GuMuTGgXuqA3nPOvyG4IzbgHaY90lq7\nN3A/wZ3qcE6oBlgAq4DjgEeNMYfhHXbKOmPMOOD3wEJr7R+C6KFX/Dg7APHAOtNa+3EArbwEnGeM\nuR74HFCG9yYQhHo+/bS6FSjEO3EelDXGmNnW2pXAMXgfcgJhjPk2sACYba3dGlQf1trVwH7xniYC\nD1trs304sNdLeOc5HwCOAv4aUB/gvV4b419vxDtEKYQvrJ7A24v4I965olMD6uMSvJFDS4wxveeu\njrHWBjbAIWjW2ieNMUcBq/H22M+21sYCaucG4B5jzIt4IxMvsda2BNQLwCJgmTGmCO/wzvIgmogf\nersJ71DX48YYgOettZcF0Y9DFgF3GWPOwjvNMD/AXr4PPGyM6QI6gNMD7MUpWiJEREScF7ZzViIi\nMgwprERExHkKKxERcZ7CSkREnKewEhER5ymsROKMMQcZY+5KsH2iMeb9+NfHGWPOj399uTHm8qw2\nKTJMhe06K5GMsda+inedSyozs9GLiOxI11nJsGKMeQs42Vq71hjzINBorT0rPiPKCmB1fKmGGXhz\n1wG8AXwRb5aDFfFtPwZ2Bw4BRgK7Ar+w1l6evUcjMnzoMKAMN78FvhT/eireTP7gTYP0oz63ux+4\n0Fp7IPAegLX2Hbw59W631v4ifrtxeEE2E7ggPrmxiAwxhZUMN78FvmSM2RdvDrhYfPG9Y/Bma8cY\nMwbYxVr7bPw+96b4eU9ba6PW2i3AFmBUxjoXGcYUVjLc/BFvdvq5eMsvPA+chDeH4Ifx2/Sw42KA\nqWb271vrfz8RGSIKKxlW4pPr/gk4Fy+sVgCXAk/1uU0d8IExpneV1r4Tm3ahgUkiWaewkuHot0CZ\ntfZveHtW4/jsysbfBi4zxqwBJvXZ/gLwLWPMOVnpVEQAjQYUEZEQ0J6ViIg4T2ElIiLOU1iJiIjz\nFFYiIuI8hZWIiDhPYSUiIs5TWImIiPP+Pw4fGKfUIv7DAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x20ba9c79518>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot('width','travel_time',data=alldata,xlim=(0,18))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 分割时间数据"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 分割出时间维度特征,包括年,月,日,小时,quarter,分钟,年第几天,年第几周,周几, \n",
    "# 以及'is_month_end',\n",
    "#  'is_month_start',\n",
    "#  'is_quarter_end',\n",
    "#  'is_quarter_start',\n",
    "#  'is_year_end',\n",
    "#  'is_year_start',\n",
    "# 甚至能否找到公路的修建时间, 可得到公路服役时间"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 同时,貌似在训练时,拿哪个时间的段的数据进行训练也是有讲究的.\n",
    "# 为了让内存不爆炸, 也要考虑分条预测?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata['year']=alldata['t0'].apply(lambda x:x.year)\n",
    "alldata['month']=alldata['t0'].apply(lambda x:x.month)\n",
    "alldata['day']=alldata['t0'].apply(lambda x:x.day)\n",
    "alldata['weekday']=alldata['t0'].apply(lambda x:x.weekday())\n",
    "alldata['hour']=alldata['t0'].apply(lambda x:x.hour)\n",
    "alldata['minute']=alldata['t0'].apply(lambda x:x.minute)\n",
    "alldata['dayofyear']=alldata['t0'].apply(lambda x:x.dayofyear)\n",
    "alldata['weekofyear']=alldata['t0'].apply(lambda x:x.weekofyear)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "alldata['is_month_end']=alldata['t0'].apply(lambda x:x.is_month_end)\n",
    "alldata['is_month_start']=alldata['t0'].apply(lambda x:x.is_month_start)\n",
    "alldata['is_quarter_end']=alldata['t0'].apply(lambda x:x.is_quarter_end)\n",
    "alldata['is_quarter_start']=alldata['t0'].apply(lambda x:x.is_quarter_start)\n",
    "alldata['is_year_end']=alldata['t0'].apply(lambda x:x.is_year_end)\n",
    "alldata['is_year_start']=alldata['t0'].apply(lambda x:x.is_year_start)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 1000 entries, 0 to 999\n",
      "Data columns (total 47 columns):\n",
      "link_ID             1000 non-null uint64\n",
      "date                1000 non-null datetime64[ns]\n",
      "t0                  1000 non-null datetime64[ns]\n",
      "t1                  1000 non-null datetime64[ns]\n",
      "delta_t             1000 non-null float64\n",
      "travel_time         1000 non-null float64\n",
      "length              1000 non-null int64\n",
      "width               1000 non-null int64\n",
      "link_class          1000 non-null int64\n",
      "in0                 962 non-null object\n",
      "in1                 238 non-null object\n",
      "in2                 79 non-null object\n",
      "in3                 33 non-null object\n",
      "out0                929 non-null object\n",
      "out1                222 non-null object\n",
      "out2                66 non-null object\n",
      "out3                32 non-null object\n",
      "in0_len             962 non-null float64\n",
      "in1_len             238 non-null float64\n",
      "in2_len             79 non-null float64\n",
      "in3_len             33 non-null float64\n",
      "out0_len            929 non-null float64\n",
      "out1_len            222 non-null float64\n",
      "out2_len            66 non-null float64\n",
      "out3_len            32 non-null float64\n",
      "in0_wid             962 non-null float64\n",
      "in1_wid             238 non-null float64\n",
      "in2_wid             79 non-null float64\n",
      "in3_wid             33 non-null float64\n",
      "out0_wid            929 non-null float64\n",
      "out1_wid            222 non-null float64\n",
      "out2_wid            66 non-null float64\n",
      "out3_wid            32 non-null float64\n",
      "year                1000 non-null int64\n",
      "month               1000 non-null int64\n",
      "day                 1000 non-null int64\n",
      "weekday             1000 non-null int64\n",
      "hour                1000 non-null int64\n",
      "minute              1000 non-null int64\n",
      "dayofyear           1000 non-null int64\n",
      "weekofyear          1000 non-null int64\n",
      "is_month_end        1000 non-null bool\n",
      "is_month_start      1000 non-null bool\n",
      "is_quarter_end      1000 non-null bool\n",
      "is_quarter_start    1000 non-null bool\n",
      "is_year_end         1000 non-null bool\n",
      "is_year_start       1000 non-null bool\n",
      "dtypes: bool(6), datetime64[ns](3), float64(18), int64(11), object(8), uint64(1)\n",
      "memory usage: 374.0+ KB\n"
     ]
    }
   ],
   "source": [
    "alldata.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>link_ID</th>\n",
       "      <th>date</th>\n",
       "      <th>t0</th>\n",
       "      <th>t1</th>\n",
       "      <th>delta_t</th>\n",
       "      <th>travel_time</th>\n",
       "      <th>length</th>\n",
       "      <th>width</th>\n",
       "      <th>link_class</th>\n",
       "      <th>in0</th>\n",
       "      <th>...</th>\n",
       "      <th>hour</th>\n",
       "      <th>minute</th>\n",
       "      <th>dayofyear</th>\n",
       "      <th>weekofyear</th>\n",
       "      <th>is_month_end</th>\n",
       "      <th>is_month_start</th>\n",
       "      <th>is_quarter_end</th>\n",
       "      <th>is_quarter_start</th>\n",
       "      <th>is_year_end</th>\n",
       "      <th>is_year_start</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>9377906285566510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 23:20:00</td>\n",
       "      <td>2016-05-21 23:22:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>17.6</td>\n",
       "      <td>143</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>4377906283759500514</td>\n",
       "      <td>...</td>\n",
       "      <td>23</td>\n",
       "      <td>20</td>\n",
       "      <td>142</td>\n",
       "      <td>20</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3377906288228510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 18:46:00</td>\n",
       "      <td>2016-05-21 18:48:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>26</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>3377906289228510514</td>\n",
       "      <td>...</td>\n",
       "      <td>18</td>\n",
       "      <td>46</td>\n",
       "      <td>142</td>\n",
       "      <td>20</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3377906284395510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 07:06:00</td>\n",
       "      <td>2016-05-21 07:08:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>10.0</td>\n",
       "      <td>43</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "      <td>3377906286918510514</td>\n",
       "      <td>...</td>\n",
       "      <td>7</td>\n",
       "      <td>6</td>\n",
       "      <td>142</td>\n",
       "      <td>20</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4377906284959500514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 14:34:00</td>\n",
       "      <td>2016-05-21 14:36:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.5</td>\n",
       "      <td>22</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>4377906282969500514</td>\n",
       "      <td>...</td>\n",
       "      <td>14</td>\n",
       "      <td>34</td>\n",
       "      <td>142</td>\n",
       "      <td>20</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9377906282776510514</td>\n",
       "      <td>2016-05-21</td>\n",
       "      <td>2016-05-21 05:04:00</td>\n",
       "      <td>2016-05-21 05:06:00</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.5</td>\n",
       "      <td>9</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>9377906283776510514</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>142</td>\n",
       "      <td>20</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 47 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "               link_ID       date                  t0                  t1  \\\n",
       "0  9377906285566510514 2016-05-21 2016-05-21 23:20:00 2016-05-21 23:22:00   \n",
       "1  3377906288228510514 2016-05-21 2016-05-21 18:46:00 2016-05-21 18:48:00   \n",
       "2  3377906284395510514 2016-05-21 2016-05-21 07:06:00 2016-05-21 07:08:00   \n",
       "3  4377906284959500514 2016-05-21 2016-05-21 14:34:00 2016-05-21 14:36:00   \n",
       "4  9377906282776510514 2016-05-21 2016-05-21 05:04:00 2016-05-21 05:06:00   \n",
       "\n",
       "   delta_t  travel_time  length  width  link_class                  in0  \\\n",
       "0      2.0         17.6     143      9           1  4377906283759500514   \n",
       "1      2.0          3.5      26     12           1  3377906289228510514   \n",
       "2      2.0         10.0      43     12           1  3377906286918510514   \n",
       "3      2.0          3.5      22      9           1  4377906282969500514   \n",
       "4      2.0          1.5       9      9           1  9377906283776510514   \n",
       "\n",
       "       ...       hour minute dayofyear weekofyear is_month_end is_month_start  \\\n",
       "0      ...         23     20       142         20        False          False   \n",
       "1      ...         18     46       142         20        False          False   \n",
       "2      ...          7      6       142         20        False          False   \n",
       "3      ...         14     34       142         20        False          False   \n",
       "4      ...          5      4       142         20        False          False   \n",
       "\n",
       "  is_quarter_end  is_quarter_start  is_year_end  is_year_start  \n",
       "0          False             False        False          False  \n",
       "1          False             False        False          False  \n",
       "2          False             False        False          False  \n",
       "3          False             False        False          False  \n",
       "4          False             False        False          False  \n",
       "\n",
       "[5 rows x 47 columns]"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "alldata.head()"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}