Bike_Share_Analysis-zh.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# 2016年美国共享单车活动快照\n",
    "\n",
    "## 目录\n",
    "- [简介](#intro)\n",
    "- [提出问题](#pose_questions)\n",
    "- [数据收集和整理](#wrangling)\n",
    "  - [浓缩骑行数据](#condensing)\n",
    "- [探索性数据分析](#eda)\n",
    "  - [统计数据](#statistics)\n",
    "  - [可视化界面](#visualizations)\n",
    "- [进行你自己的分析](#eda_continued)\n",
    "- [总结](#conclusions)\n",
    "\n",
    "<a id='intro'></a>\n",
    "## 简介\n",
    "\n",
    "> **小贴士**: 像这样的引用部分旨在为学员提供实用指导，帮助学员了解并使用 Jupyter notebook\n",
    "\n",
    "过去十年里，世界各城市的共享单车系统不仅不断增多，也越来越受人们欢迎。用户可借助共享单车系统来租赁单车，进行通常在 30 分钟以内的短途骑行。由于信息科技的崛起，用户能够方便地找到所用系统的停车点，实现单车解锁或归还操作。不仅如此，信息科技还带来了大量数据，有了这些数据，我们就能研究共享单车系统的使用情况。\n",
    "\n",
    "在本项目中，你将进行探索性数据分析。你所需分析的数据源自 [Motivate](https://www.motivateco.com/)，这是一家为美国诸多大城市提供共享单车系统的供应商。你将比较三大城市：纽约市、芝加哥和华盛顿特区的系统使用情况，并观察每个系统的长期注册用户和短期临时用户之间有无区别。\n",
    "\n",
    "<a id='pose_questions'></a>\n",
    "## 提出问题\n",
    "\n",
    "在查看共享单车数据之前，你得先明确自己想了解共享单车数据的哪些方面，并就此提出问题。打个比方，假如你是 Motivate 的员工，为了做出更明智的商业决策，你会想了解什么信息呢？假如你是共享单车服务的对象，那么什么样的因素会影响你对这类服务的使用偏好呢？\n",
    "\n",
    "**问题 1**: 写出至少两个你觉得能用数据解决的共享单车相关问题。\n",
    "\n",
    "**答案**：1.根据数据决定单车的投放点，比如某个地点在某个时间段为单车使用的高峰，那么就可以在合适的时间将大量的单车投放到该地点，提升用户使用率及用户体验\n",
    "2.根据用户的历史行为作出判定，如果该用户使用的单车损毁率高于一个阈值的话那么这个用户可能存在蓄意破坏的行为，可以对此用户进行警告，警告无效后可以将此用户拉入黑名单，不允许再使用单车\n",
    "\n",
    "\n",
    "\n",
    "<a id='wrangling'></a>\n",
    "## 数据收集与整理\n",
    "\n",
    "现在是时候收集并探索数据了。在本项目中，我们将重点关注所选城市：纽约市、芝加哥、华盛顿特区 2016 年的个人骑行记录。每座城市均有相关页面供我们随时下载骑行数据：\n",
    "\n",
    "- 纽约市(Citi Bike): [链接](https://www.citibikenyc.com/system-data)\n",
    "- 芝加哥 (Divvy): [链接](https://www.divvybikes.com/system-data)\n",
    "- 华盛顿特区 (Capital Bikeshare): [链接](https://www.capitalbikeshare.com/system-data)\n",
    "\n",
    "访问上述网页，你就会发现各城市提供数据的方式有所不同。芝加哥每两年更新一次数据，华盛顿特区则每个季度更新一次，纽约市则是每月一更。**不过你无需亲自下载数据。** 我们已为你收集好了数据并存放到项目文件的 `/data/` 文件夹中了，虽然各城市 2016 年的原始数据散布在多个文件中，但 `/data/` 文件夹里的文件已分别将当年各城市的所有骑行数据整合到一个文件中了。我们已为你完成了一些数据整理操作，处理了各城市数据时间戳格式不一致的问题。除此之外，我们还移除了原始数据 2% 的随机样本，以减少探索障碍。\n",
    "\n",
    "**问题 2**: 尽管如此，我们要研究的数据还有很多，因此，我们最好先观察各目标分析城市的某一则记录。运行下方第一个代码框，加载一些分析所需的包和函数，然后完成第二个代码框，使其分别输出各城市的第一条骑行记录（也即每个数据文件的第二行）。\n",
    "\n",
    "> **小贴士**: 运行代码框的方法与编辑上方的 Markdown 框的格式类似，你只需点击代码框，按下键盘快捷键 **Shift** + **Enter** 或 **Shift** + **Return** ，或者你也可先选择代码框，然后点击工具栏的 **运行** 按钮来运行代码。运行代码框时，相应单元左侧的信息会出现星号，即 `In [*]:`，若代码执行完毕，星号则会变为某个数字，如 `In [1]`。如果代码运行后有输出结果，输出将会以 `Out [1]:` 的形式出现，其中的数字将与 \"In\" 中的数字相对应。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## 导入所有必要的包和函数。\n",
    "import csv # 读写 csv 文件\n",
    "from datetime import datetime # 日期解析操作\n",
    "from pprint import pprint # 用于输出字典等数据结构\n",
    "                          # 这比 base print 函数要好用。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "City: NYC\n",
      "OrderedDict([('tripduration', '839'),\n",
      "             ('starttime', '1/1/2016 00:09:55'),\n",
      "             ('stoptime', '1/1/2016 00:23:54'),\n",
      "             ('start station id', '532'),\n",
      "             ('start station name', 'S 5 Pl & S 4 St'),\n",
      "             ('start station latitude', '40.710451'),\n",
      "             ('start station longitude', '-73.960876'),\n",
      "             ('end station id', '401'),\n",
      "             ('end station name', 'Allen St & Rivington St'),\n",
      "             ('end station latitude', '40.72019576'),\n",
      "             ('end station longitude', '-73.98997825'),\n",
      "             ('bikeid', '17109'),\n",
      "             ('usertype', 'Customer'),\n",
      "             ('birth year', ''),\n",
      "             ('gender', '0')])\n",
      "\n",
      "City: Chicago\n",
      "OrderedDict([('trip_id', '9080545'),\n",
      "             ('starttime', '3/31/2016 23:30'),\n",
      "             ('stoptime', '3/31/2016 23:46'),\n",
      "             ('bikeid', '2295'),\n",
      "             ('tripduration', '926'),\n",
      "             ('from_station_id', '156'),\n",
      "             ('from_station_name', 'Clark St & Wellington Ave'),\n",
      "             ('to_station_id', '166'),\n",
      "             ('to_station_name', 'Ashland Ave & Wrightwood Ave'),\n",
      "             ('usertype', 'Subscriber'),\n",
      "             ('gender', 'Male'),\n",
      "             ('birthyear', '1990')])\n",
      "\n",
      "City: Washington\n",
      "OrderedDict([('Duration (ms)', '427387'),\n",
      "             ('Start date', '3/31/2016 22:57'),\n",
      "             ('End date', '3/31/2016 23:04'),\n",
      "             ('Start station number', '31602'),\n",
      "             ('Start station', 'Park Rd & Holmead Pl NW'),\n",
      "             ('End station number', '31207'),\n",
      "             ('End station', 'Georgia Ave and Fairmont St NW'),\n",
      "             ('Bike number', 'W20842'),\n",
      "             ('Member Type', 'Registered')])\n"
     ]
    }
   ],
   "source": [
    "def print_first_point(filename):\n",
    "    \"\"\"\n",
    "    本函数会输出并返回指定的 csv 文件 （含页眉行）的第一个数据点（即文件的第二行）。\n",
    "    \"\"\"\n",
    "    # 输出城市名以供参考\n",
    "    city = filename.split('-')[0].split('/')[-1]\n",
    "    print('\\nCity: {}'.format(city))\n",
    "    \n",
    "    with open(filename, 'r') as f_in:\n",
    "        ## 待办：用 csv 库来设置一个 DictReader 对象。##\n",
    "        ## 见 https://docs.python.org/3/library/csv.html           ##\n",
    "        trip_reader = csv.DictReader(f_in)\n",
    "        \n",
    "        ## 待办：对 DictReader 对象使用函数     ##\n",
    "        ## 从而读取数据文件的第一条骑行记录并将其存储为一个变量     ##\n",
    "        ## 见 https://docs.python.org/3/library/csv.html#reader-objects ##\n",
    "        first_trip = trip_reader.__next__()\n",
    "        \n",
    "        ## 待办：用 pprint 库来输出第一条骑行记录。 ##\n",
    "        ## 见 https://docs.python.org/3/library/pprint.html     ##\n",
    "        pprint(first_trip)\n",
    "    # 输出城市名和第一条骑行记录以备测试\n",
    "    return (city, first_trip)\n",
    "\n",
    "# 各城市的文件列表\n",
    "data_files = ['./data/NYC-CitiBike-2016.csv',\n",
    "              './data/Chicago-Divvy-2016.csv',\n",
    "              './data/Washington-CapitalBikeshare-2016.csv',]\n",
    "\n",
    "# 输出各文件的第一条骑行记录，并将其储存在字典中\n",
    "example_trips = {}\n",
    "for data_file in data_files:\n",
    "    city, first_trip = print_first_point(data_file)\n",
    "    example_trips[city] = first_trip"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "要是一切都填写正确的话，你就能看到下方输出了每座城市的名字（解析数据文件名而得），而其中第一条骑行记录也已解析成字典的格式。等你设置好`DictReader`对象，数据文件的第一行通常就会被处理为列名，文件其余的行则将以这些列名作为关键字，这是因为每一行都生成了一个字典对象。\n",
    "\n",
    "这些可读性很强的标签对我们访问大量数据十分实用，有了这些，我们就无需使用数字索引，比如，如果我们有一条骑行记录储存在变量 `row`里，那就可以用 `row['duration']` 来获得骑行时长，无需使用 `row[0]`。\n",
    "\n",
    "<a id='condensing'></a>\n",
    "### 浓缩骑行数据\n",
    "\n",
    "你应该能从上方的输出观察到每座城市提供的信息是有差别的，就算是信息相同的部分，列名和格式有时也不尽相同。为了尽可能方便真正的探索操作，我们应当对数据进行修剪和清理。清理数据以保证各城市数据格式一致，修剪数据则能让我们把重心完全放到最关心的那部分数据上，从而减小探索难度。\n",
    "\n",
    "你将生成新的数据文件，该文件需包括你感兴趣的五个骑行数据值：骑行时长、起始月份、起始小时、周几以及用户类型。每座城市的这五个值都可能需要额外的整理操作，这取决于各城市的数据差异：\n",
    "\n",
    "- **时长**：该值可能以秒数（纽约市、芝加哥）或毫秒数（华盛顿特区）表示，但就分析而言，分析单位为分钟更佳，所以如果能将所有骑行时长都以分钟表示就好了。\n",
    "- **月份**, **小时**, **周几**：季节、一天的什么时候以及是否周末都有可能影响骑客量，请用骑行起始时间来获取这些值。纽约市数据的时间戳记录了秒数，而华盛顿特区和芝加哥的并没有。 此处使用 [`datetime`](https://docs.python.org/3/library/datetime.html) 包应大有帮助，该包可实现必要的转换操作。\n",
    "- **用户类型**： 共享单车系统的注册用户可能与临时用户有不同的使用模式。华盛顿特区将其用户分为两种：‘注册用户’ ——这类用户买了较长时期的会员，如年度会员或月度会员；‘临时用户’——这类用户所购买的骑行时间较短，如只有 24 小时或 3 天。纽约市和芝加哥的数据则用 ‘会员’ 和 ‘散客’ 来区分这两者。为了保证数据的一致性，你需要修改华盛顿特区的标签，使其与另外两座城市的标签相同。\n",
    "\n",
    "\n",
    "**问题 3a**：完成下方代码框中的辅助函数，完成上述各项清理任务。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def duration_in_mins(datum, city):\n",
    "    \"\"\"\n",
    "    将一个字典作为输入，该字典需包含一条骑行记录（数据）\n",
    "    及记录城市（城市）的信息，返回该骑行的时长，使该时长以分钟为单位。\n",
    "    \n",
    "    记住，华盛顿特区是以毫秒作为计量单位的，而芝加哥和纽约市则\n",
    "    以秒数作为单位。\n",
    "    \n",
    "    提示：csv 模块会将所有数据读取为字符串，包括数值，\n",
    "    所以转换单位时，你需要用一个函数来将字符串转换为合适的数值类型。\n",
    "    见 https://docs.python.org/3/library/functions.html\n",
    "    \"\"\"\n",
    "    \n",
    "    # 请在此处写出代码\n",
    "    duration = 0.0\n",
    "    if city == 'Washington':\n",
    "        duration = float(datum['Duration (ms)'])\n",
    "        duration = duration / 1000 / 60\n",
    "    else:\n",
    "        duration = float(datum['tripduration'])\n",
    "        duration = duration / 60\n",
    "    return duration\n",
    "\n",
    "\n",
    "# 测试代码是否奏效，若所有断言都没问题，则不应有输出出现。\n",
    "# 至于字典 `example_trips` \n",
    "# 则是在你输出每个数据源文件的第一条骑行数据时生成的。\n",
    "tests = {'NYC': 13.9833,\n",
    "         'Chicago': 15.4333,\n",
    "         'Washington': 7.1231}\n",
    "\n",
    "for city in tests:\n",
    "    assert abs(duration_in_mins(example_trips[city], city) - tests[city]) < .001"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def time_of_trip(datum, city):\n",
    "    \"\"\"\n",
    "    将一个字典作为输入，该字典需包含一条骑行记录（数据）\n",
    "    及记录城市（城市）的信息，返回该骑行进行的月份、小时及周几这三个值。\n",
    "    \n",
    "    \n",
    "    记住，纽约市以秒为单位，华盛顿特区和芝加哥则不然。\n",
    "    \n",
    "    提示：你需要用 datetime 模块来将原始日期字符串解析为\n",
    "    方便提取目的信息的格式。\n",
    "    见 https://docs.python.org/3/library/datetime.html#strftime-and-strptime-behavior\n",
    "    \"\"\"\n",
    "    \n",
    "    # 请在此处写出代码\n",
    "    month = 0\n",
    "    hour = 0\n",
    "    day_of_week = ''\n",
    "    key = ''\n",
    "    fmt = ''\n",
    "    if city == 'Washington':\n",
    "        key = 'Start date'\n",
    "        fmt = '%m/%d/%Y %H:%M'\n",
    "    elif city == 'NYC':\n",
    "        key = 'starttime'\n",
    "        fmt = '%m/%d/%Y %H:%M:%S'\n",
    "    elif city == 'Chicago':\n",
    "        key = 'starttime'\n",
    "        fmt = '%m/%d/%Y %H:%M'\n",
    "    date = datetime.strptime(datum[key],fmt)\n",
    "    week = ['Monday','Tuesday','Wednesday','Thursday','Friday','Saturday','Sunday']\n",
    "    month = date.month\n",
    "    hour = date.hour\n",
    "    day_of_week = week[date.weekday()]\n",
    "    return (month, hour, day_of_week)\n",
    "\n",
    "\n",
    "# 测试代码是否奏效，若所有断言都没问题，则不应有输出出现。\n",
    "# 至于字典 `example_trips`\n",
    "# 则是在你输出每个数据源文件的第一条骑行数据时生成的。\n",
    "tests = {'NYC': (1, 0, 'Friday'),\n",
    "         'Chicago': (3, 23, 'Thursday'),\n",
    "         'Washington': (3, 22, 'Thursday')}\n",
    "\n",
    "for city in tests:\n",
    "    assert time_of_trip(example_trips[city], city) == tests[city]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def type_of_user(datum, city):\n",
    "    \"\"\"\n",
    "    将一个字典作为输入，该字典需包含一条骑行记录（数据）\n",
    "    及记录城市（城市）的信息，返回进行该骑行的系统用户类型。\n",
    "    \n",
    "    \n",
    "    记住，华盛顿特区的类名与芝加哥和纽约市的不同。\n",
    "    \n",
    "    \"\"\"\n",
    "    \n",
    "    # 请在此处写出代码\n",
    "    user_type = \"\"\n",
    "    if city == 'Washington':\n",
    "        user_type = 'Subscriber' if datum['Member Type'] == 'Registered' else 'Customer'\n",
    "    else:\n",
    "        user_type = datum['usertype']\n",
    "    return user_type\n",
    "\n",
    "\n",
    "# 测试代码是否奏效，若所有断言都没问题，则不应有输出出现。\n",
    "# 至于字典 `example_trips`\n",
    "# 则是在你输出每个数据源文件的第一条骑行数据时生成的。\n",
    "tests = {'NYC': 'Customer',\n",
    "         'Chicago': 'Subscriber',\n",
    "         'Washington': 'Subscriber'}\n",
    "\n",
    "for city in tests:\n",
    "    assert type_of_user(example_trips[city], city) == tests[city]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**问题 3b**: 现在，使用你在上方编写的辅助函数来为每座城市生成数据浓缩文件，该文件将只包括上述数据区。在 `/examples/` 文件夹里，你将看到转换前后的数据文件示例，该示例来自[旧金山湾区共享单车](http://www.bayareabikeshare.com/open-data)。请确保你的输出结果与示例文件具备一致的格式。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def condense_data(in_file, out_file, city):\n",
    "    \"\"\"\n",
    "    本函数会从指定的输入文件中提取全部数据\n",
    "    并在指定的输出文件中写出浓缩数据。\n",
    "    城市参数决定输入文件的解析方式。\n",
    "    \n",
    "    提示：参考下框以明确参数结构！\n",
    "    \"\"\"\n",
    "    \n",
    "    with open(out_file, 'w') as f_out, open(in_file, 'r') as f_in:\n",
    "        # 设置 csv DictWriter 对象——该对象需将第一列列名\n",
    "        # 作为 \"fieldnames\" 参数\n",
    "        out_colnames = ['duration', 'month', 'hour', 'day_of_week', 'user_type']        \n",
    "        trip_writer = csv.DictWriter(f_out, fieldnames = out_colnames)\n",
    "        trip_writer.writeheader()\n",
    "        \n",
    "        ## 待办：设置 csv DictReader 对象##\n",
    "        trip_reader = csv.DictReader(f_in)\n",
    "\n",
    "        # 收集并处理每行的数据\n",
    "        week = ['Monday','Tuesday','Wednesday','Thursday','Friday','Saturday','Sunday']\n",
    "        for row in trip_reader:\n",
    "            # 设置一个字典来存储清理和修剪后的数据点的值\n",
    "            new_point = {}\n",
    "\n",
    "            ## 待办：使用辅助函数来从原始数据字典中获取清理数据##\n",
    "            ## 注意字典 new_point 的关键词应与 ##\n",
    "            ## 上述 DictWriter 对象设置的列名一致。        ##\n",
    "            new_point['duration'] = duration_in_mins(row,city)\n",
    "            time = time_of_trip(row,city)\n",
    "            new_point['month'] = time[0]\n",
    "            new_point['hour'] = time[1]\n",
    "            new_point['day_of_week'] = time[2]\n",
    "            new_point['user_type'] = type_of_user(row,city)\n",
    "            ## 待办：在输出文件中写出处理后的信息。##\n",
    "            ## 见 https://docs.python.org/3/library/csv.html#writer-objects ##\n",
    "            trip_writer.writerow(new_point)\n",
    "            "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "City: Washington\n",
      "OrderedDict([('duration', '7.123116666666666'),\n",
      "             ('month', '3'),\n",
      "             ('hour', '22'),\n",
      "             ('day_of_week', 'Thursday'),\n",
      "             ('user_type', 'Subscriber')])\n",
      "\n",
      "City: Chicago\n",
      "OrderedDict([('duration', '15.433333333333334'),\n",
      "             ('month', '3'),\n",
      "             ('hour', '23'),\n",
      "             ('day_of_week', 'Thursday'),\n",
      "             ('user_type', 'Subscriber')])\n",
      "\n",
      "City: NYC\n",
      "OrderedDict([('duration', '13.983333333333333'),\n",
      "             ('month', '1'),\n",
      "             ('hour', '0'),\n",
      "             ('day_of_week', 'Friday'),\n",
      "             ('user_type', 'Customer')])\n",
      "finish\n"
     ]
    }
   ],
   "source": [
    "# 运行下框以测试效果\n",
    "city_info = {'Washington': {'in_file': './data/Washington-CapitalBikeshare-2016.csv',\n",
    "                            'out_file': './data/Washington-2016-Summary.csv'},\n",
    "             'Chicago': {'in_file': './data/Chicago-Divvy-2016.csv',\n",
    "                         'out_file': './data/Chicago-2016-Summary.csv'},\n",
    "             'NYC': {'in_file': './data/NYC-CitiBike-2016.csv',\n",
    "                     'out_file': './data/NYC-2016-Summary.csv'}}\n",
    "\n",
    "for city, filenames in city_info.items():\n",
    "    condense_data(filenames['in_file'], filenames['out_file'], city)\n",
    "    print_first_point(filenames['out_file'])\n",
    "print('finish')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **小贴士**：对 Jupyter Notebook 进行保存，运行代码块所得的输出结果也会得到保存，不过，一旦开始了新节，工作空间的状态就会被重置，所以你得运行上一节涉及的所有必要代码块，重建变量和函数，方能找回上次离开的进度。\n",
    "\n",
    "<a id='eda'></a>\n",
    "## 探索性数据分析\n",
    "\n",
    "现在你已经收集并整理了数据，可以开始探索数据了。在本节中，你将编写一些代码来对数据进行描述统计。我们还将为你介绍 `matplotlib` 库，以便你为数据创建一些基本直方图。\n",
    "\n",
    "<a id='statistics'></a>\n",
    "### 统计数据\n",
    "\n",
    "首先，我们来进行一些基本统计。下方的第一个框里包括了一个函数，该函数会用 csv 模块来对给定的数据文件进行迭代处理，并返回会员和散客的骑行次数。第二个框则会运行 `/examples/` 文件夹里湾区示例数据中的函数。修改下框，回答下列问题。\n",
    "\n",
    "**问题 4a**：哪座城市的骑行次数最多？哪座城市会员进行的骑行次数占比最高？哪座城市散客进行的骑行次数占比最高？\n",
    "\n",
    "**答案**：纽约骑行次数最多；纽约会员骑行次数占比最高；芝加哥散客骑行次数最高"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def number_of_trips(filename):\n",
    "    \"\"\"\n",
    "    本函数会读取一个骑行数据文件，分别报告\n",
    "    会员、散客和所有系统用户的骑行次数。\n",
    "    \"\"\"\n",
    "    with open(filename, 'r') as f_in:\n",
    "        # 设置 csv reader 对象\n",
    "        reader = csv.DictReader(f_in)\n",
    "        \n",
    "        # 初始化计数变量\n",
    "        n_subscribers = 0\n",
    "        n_customers = 0\n",
    "        n_total = 0\n",
    "        \n",
    "        # 计算骑行类型\n",
    "        for row in reader:\n",
    "            if row['user_type'] == 'Subscriber':\n",
    "                n_subscribers += 1\n",
    "            else:\n",
    "                n_customers += 1\n",
    "\n",
    "        \n",
    "        # 统计骑行总次数\n",
    "        n_total = n_subscribers + n_customers\n",
    "        sub_rate = n_customers / float(n_total)\n",
    "        # 将结果作为数组返回出来\n",
    "        return(n_subscribers, n_customers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(51753, 14573)\n"
     ]
    }
   ],
   "source": [
    "## 修改此框及上框，回答问题 4a。##\n",
    "## 记得运行你在问题 3 中创建的数据文件清理函数。     ##\n",
    "\n",
    "data_file = './data/Washington-2016-Summary.csv'\n",
    "print(number_of_trips(data_file))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> **小贴士**：要在 notebook 里添加框，你可以用上方菜单栏里的  \"在上方插入框\" 和 \"在下方插入框\" 功能，也可以用工具栏的相关图标来添加新框，那里还有其它图标可在文件中上移或下移框。新框默认为代码框。你还可以在框菜单或工具栏下拉菜单中为选中框指定类型（如代码框或 Markdown 框）。\n",
    "\n",
    "现在，你需要写下自己的代码，继续研究数据的属性。\n",
    "\n",
    "**问题 4b**：共享单车系统是为短途骑行者而设的。大多数时候，用户在 30 分钟内骑行无需额外付费，若是超过该时长，则需支付超时费用。那么，各城市的平均骑行时长是多少？各城市骑行时长超过 30 分钟的比例又是多少？\n",
    "\n",
    "**答案**：华盛顿平均时长18.4分钟，超过30分钟比例为0.1；芝加哥平均时长16.1分钟，超过30分钟比例为0.08；纽约平均时长15.3分钟，超过30分钟比例为0.07"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('16.6', '0.08')\n"
     ]
    }
   ],
   "source": [
    "## 使用本框及新框来回答问题 4b。               ##\n",
    "##                                                                      ##\n",
    "## 提示：csv 模块会将所有数据读取为字符串，包括数值。 ##\n",
    "## 因此，在统计数据之前，你需要用函数将字符串转换为      ##\n",
    "## 合适的数值类型。         ##\n",
    "## 小贴士：在湾区示例数据中，平均骑行时长为 14 分钟，##\n",
    "## 骑行时长多于 30 分钟的数据占比 3.5%。                      ##\n",
    "def avg_ride_time(filename):\n",
    "    with open(filename,'r') as file:\n",
    "        count = 0\n",
    "        over30 = 0.0\n",
    "        total_time = 0.0\n",
    "        reader = csv.DictReader(file)\n",
    "        for row in reader:\n",
    "            count += 1\n",
    "            time = float(row['duration'])\n",
    "            total_time += time\n",
    "            if time > 30:\n",
    "                over30 += 1\n",
    "        \n",
    "    return ('%.1f' % (total_time / count) ,'%.2f' % (over30 / count))\n",
    "data_file = './data/Chicago-2016-Summary.csv'\n",
    "print(avg_ride_time(data_file))\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**问题 4c**：从骑客量的角度更深入地挖掘骑行时长问题。选择一座城市，研究该城市哪种系统用户的平均骑行时间更长？是会员还是散客？\n",
    "\n",
    "**答案**：芝加哥会员平均骑行11.6分钟，散客平均骑行30.5分钟；纽约会员平均骑行13.2分钟，散客平均骑行32.3分钟；华盛顿会员平均骑行12分钟，散客平均骑行41.2分钟"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('12.5', '41.7')\n"
     ]
    }
   ],
   "source": [
    "## 使用本框及新框来回答问题 4c。##\n",
    "## 如果你还没这么做过，你可以考虑修改之前的代码   ##\n",
    "## 利用一些可重复利用的函数。                            ##\n",
    "##                                                                     ##\n",
    "## 小贴士：在海湾示例数据中，你应该发现    ##\n",
    "## 会员平均骑行时长为 9.5 分钟，散客平均骑行时长则为##\n",
    "## 54.6 分钟，其它城市区别也这么大吗？     ##\n",
    "##                                                ##\n",
    "def time_of_rides(filename):\n",
    "    with open(filename) as file:\n",
    "        reader = csv.DictReader(file)\n",
    "        sub_time = 0\n",
    "        cus_time = 0\n",
    "        for row in reader:\n",
    "            if row['user_type'] == 'Subscriber':\n",
    "                sub_time += float(row['duration'])\n",
    "            else:\n",
    "                cus_time += float(row['duration'])\n",
    "    return (sub_time,cus_time)\n",
    "data_file = './data/Washington-2016-Summary.csv'\n",
    "sub_time,cus_time = time_of_rides(data_file)\n",
    "sub_number,cus_number = number_of_trips(data_file)\n",
    "print(('%.1f' % (sub_time / sub_number),'%.1f' % (cus_time / cus_number)))       \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='visualizations'></a>\n",
    "### 可视化界面\n",
    "\n",
    "你统计出来的最后一组值应该会出现相当有趣的结果：会员平均骑行时长远低于 30 分钟，而散客的平均骑行时长则的的确确 _超过_了 30 分钟！要是能观察骑行时长的分布，应该会很有趣，为此，我们需要引入一个新库，即 `matplotlib`。请运行下框，加载该库以生成示例图。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAW4AAAEWCAYAAABG030jAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107ff7d68>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 加载库\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "# 这个'咒语'能展示图形。\n",
    "# 内联 notebook，详见：\n",
    "# http://ipython.readthedocs.io/en/stable/interactive/magics.html\n",
    "%matplotlib inline \n",
    "\n",
    "# 直方图示例，数据来自湾区样本\n",
    "data = [ 7.65,  8.92,  7.42,  5.50, 16.17,  4.20,  8.98,  9.62, 11.48, 14.33,\n",
    "        19.02, 21.53,  3.90,  7.97,  2.62,  2.67,  3.08, 14.40, 12.90,  7.83,\n",
    "        25.12,  8.30,  4.93, 12.43, 10.60,  6.17, 10.88,  4.78, 15.15,  3.53,\n",
    "         9.43, 13.32, 11.72,  9.85,  5.22, 15.10,  3.95,  3.17,  8.78,  1.88,\n",
    "         4.55, 12.68, 12.38,  9.78,  7.63,  6.45, 17.38, 11.90, 11.52,  8.63,]\n",
    "plt.hist(data)\n",
    "plt.title('Distribution of Trip Durations')\n",
    "plt.xlabel('Duration (m)')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "在上框中，我们收集了五十个骑行时长数据，制成列表，并将该列表作为 `.hist()` 函数的第一个参数，该函数会进行统计并生成绘图对象，以便生成直方图，但我们需要执行 `.show()` 函数才能呈现统计图。`.title()` 和 `.xlabel()` 函数则提供一些标签，以便说明统计图的背景。\n",
    "\n",
    "现在你可以使用这些函数，为你在问题 4c 所选城市的骑行时长创建直方图。请先不要将会员和散客的统计分开，你现在只需统计所有骑行时长并生成统计图即可。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "66326\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEWCAYAAACNJFuYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAHdNJREFUeJzt3XucXGWd5/HP14RLuIQkJmBIMgaW\njGNkhlsIcZn1AmwIiATnBTth0bRM3LyWQdfxshoGRxRBQXd0ZFU0CzEBuZqRITJgJgYzjjMS0sgl\nhIhpQyBtgDQGkDsEfvvHeQoO9VR3VXV3ujr09/161avOeZ7nPOep0931rXOp04oIzMzMyt7U6gGY\nmdng43AwM7OMw8HMzDIOBzMzyzgczMws43AwM7OMw8G6Jem7kv6un/r6I0lPSxqW5ldJ+kh/9J36\nu0VSW3/118R6L5D0mKRH+qm/+yX9l/7oq1XeCK/BQP6ew9AkaROwH7AdeBm4D7gCWBgRr/Sir49E\nxE+bWGYV8IOIuKyZdaVlvwAcFBEfbHbZ/iRpEvAb4K0RsbWq7gzge2l2GLAb8GylPiL26uexDAde\nSusI4HngLuB7EfHD/lxX1Xp/AHRExBd21DqsNbznMLS9PyL2Bt4KXAR8Fri8v1eS3rjeiN4K/L46\nGAAi4qqI2CuFwAnAlsp8rWDox230jtT/nwA/AC6VdG5vOnoD/9ysERHhxxB8AJuA46rKpgOvAAen\n+cXABWl6LHAT8ASwDfg3ig8XV6ZlngOeBj4DTKb49DoPeAj4ealseOpvFfAV4HbgSeBGYEyqew/Q\nWWu8wCzgRYpPyU8Dd5f6+0iafhPwOeBBYCvFHtE+qa4yjrY0tseAc3vYTvuk5btSf59L/R+XXvMr\naRyLe+gjez2pvBP438Ba4MVS2XvS9AXAdcAPgaeAduBPu1nH8PS6JleVz0njHFXdf2kdi9P0QamP\nM9O2uTW91qXAI+lnvwp4e2r/1+nn8GLaBjfUeA27A5cADwO/A74O7Jrqjks/18+k7bsFmFsa20nA\n+vTaO4FPtPrvZig9vOdgr4qI2yn+CGsdL/5UqhtHcTjqb4tF4kMUbyTvj+JT8VdLy7wbeDtwfDer\nnAv8FbA/xeGtSxoY40+ALwPXpfUdUqPZh9PjvcCBwF7At6ra/DnwNuBY4POS3t7NKv8vRUAcmF7P\nXODMKA6hlfcIPlxv7N2Yk/rZp5v6vwCuBsZQvEnf0OQn+n+iOKR1ZBPLvItiz+N9af4mYArwFuBe\nig8ERMR3KMLry2kbfKBGX58HpgF/BhwGHA2cU6qfCIyg+B34nxR7OiNT3feBeVHs3f4Z8K9NvAbr\nI4eDVdtC8UZU7SVgPMXx9Zci4t8ifbzrwRci4pmIeK6b+isj4t6IeAb4O+C/VU5Y99EZwNcjYmNE\nPE3xZjSn6k31ixHxXETcDdwNZCGTxvKXwDkR8VREbAL+HvhQP4yx4psR0dnDNlodETdExEvA14CR\nNPFGHxHPU+zp1fqZdue8iHg2bZ9XImJxev3PA18AjpC0Z4N9nUHxe9AVxeG383n99nueYu/0pYhY\nBrwA/HGqewmYKmnviNgWEb9q4jVYHzkcrNoEijeTal8DOoB/kbRR0oIG+trcRP2DwC4Uh6/6av/U\nX7nv4RR7PBXlq4uepdi7qDYW2LVGXxP6YYwVDW+jiHiZ4tDM/o12Lml3imCo9TOtu05JwyR9Nf3M\n/0DxOwCN/5zG0/P2eyy9roryz+IDwMnAQ+nqtqOaeA3WRw4He5WkIyn+cH9RXZc+OX4qIg4E3g98\nUtKxlepuuqy3ZzGpNP1HFJ8UHwOeAfYojWsYxeGsRvvdQnGyuNz3duDROstVeyyNqbqv3zXZT08a\n3kaS3kTx89nSRP+nUHwaX5PmX7dtKQ4VvX5Ar98jnAucCBxDcejroMpwKs3rrP9hern9ImJ1RJwM\n7EtxaOvaRpaz/uFwMCSNlHQSxR/fDyJibY02J0k6SJKAP1Bc/lr5xPcoxTH5Zn1Q0lRJe1Acblia\nPkX+Bthd0vsk7UJxEni30nKPApPTm2Ut1wCfkHSApL147RzF9mYGl8ZyPXChpL0lvRX4JMVVQANl\nuqTZaTt8muLk7Jo6yyDpzZI+RHHO5CsR8USquot0iE3SdIpzGj3ZmyJcfk8RKhdW1df72V9DcU5n\nrKRxFIcP624/SSMk/XdJI9Mhtad47ffNBoDDYWj7saSnKA4jnEtxJcmZ3bSdAvyU4qqUXwLfiYhV\nqe4rwOckPSHp002s/0qKK6Ieobiq5X8BRMSTFFfCXEbxKfMZipPhFZXr9n8vqdZx6EWp758DD1Ac\n1/5YE+Mq+1ha/0aKPaqrU/8D5QbggxSHhf4S+Is6IbdO0tPABoqf5cci4vxS/bkUJ5ufoHijvrrO\n+r9PsaeyBVgH/EdV/WXAIZIel7S0xvJfpDinsxa4B1hN8fvSiDbgwXQ4ax79e67H6vCX4MwGKUkX\nABP7cCWUWa95z8HMzDIOBzMzy/iwkpmZZbznYGZmmZ32xlpjx46NyZMnt3oYZmY7jTvuuOOxiBhX\nv+VOHA6TJ0+mvb291cMwM9tpSHqwfquCDyuZmVnG4WBmZhmHg5mZZRwOZmaWcTiYmVnG4WBmZhmH\ng5mZZRwOZmaWcTiYmVlmp/2GdF9MXvDPLVnvpove15L1mpk1y3sOZmaWcTiYmVnG4WBmZhmHg5mZ\nZRwOZmaWcTiYmVnG4WBmZhmHg5mZZRwOZmaWcTiYmVnG4WBmZhmHg5mZZRwOZmaWcTiYmVmmoXCQ\nNErSUkm/lrRe0jsljZG0QtKG9Dw6tZWkSyR1SLpH0uGlftpS+w2S2krlR0ham5a5RJL6/6WamVmj\nGt1z+Cbwk4j4E+AQYD2wAFgZEVOAlWke4ARgSnrMBy4FkDQGOA84CpgOnFcJlNRmfmm5WX17WWZm\n1hd1w0HSSOBdwOUAEfFiRDwBzAaWpGZLgFPS9GzgiijcBoySNB44HlgREdsi4nFgBTAr1Y2MiF9G\nRABXlPoyM7MWaGTP4UCgC/i+pDslXSZpT2C/iHgYID3vm9pPADaXlu9MZT2Vd9Yoz0iaL6ldUntX\nV1cDQzczs95oJByGA4cDl0bEYcAzvHYIqZZa5wuiF+V5YcTCiJgWEdPGjRvX86jNzKzXGgmHTqAz\nIlan+aUUYfFoOiREet5aaj+ptPxEYEud8ok1ys3MrEXqhkNEPAJslvS2VHQscB+wDKhccdQG3Jim\nlwFz01VLM4An02Gn5cBMSaPTieiZwPJU95SkGekqpbmlvszMrAWGN9juY8BVknYFNgJnUgTL9ZLm\nAQ8Bp6W2NwMnAh3As6ktEbFN0peANand+RGxLU2fBSwGRgC3pIeZmbVIQ+EQEXcB02pUHVujbQBn\nd9PPImBRjfJ24OBGxmJmZjuevyFtZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmYZh4OZmWUcDmZmlnE4\nmJlZxuFgZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmYZh4OZmWUc\nDmZmlnE4mJlZxuFgZmYZh4OZmWUaCgdJmyStlXSXpPZUNkbSCkkb0vPoVC5Jl0jqkHSPpMNL/bSl\n9hsktZXKj0j9d6Rl1d8v1MzMGtfMnsN7I+LQiJiW5hcAKyNiCrAyzQOcAExJj/nApVCECXAecBQw\nHTivEiipzfzScrN6/YrMzKzP+nJYaTawJE0vAU4plV8RhduAUZLGA8cDKyJiW0Q8DqwAZqW6kRHx\ny4gI4IpSX2Zm1gKNhkMA/yLpDknzU9l+EfEwQHreN5VPADaXlu1MZT2Vd9Yoz0iaL6ldUntXV1eD\nQzczs2YNb7Dd0RGxRdK+wApJv+6hba3zBdGL8rwwYiGwEGDatGk125iZWd81tOcQEVvS81bgBopz\nBo+mQ0Kk562peScwqbT4RGBLnfKJNcrNzKxF6oaDpD0l7V2ZBmYC9wLLgMoVR23AjWl6GTA3XbU0\nA3gyHXZaDsyUNDqdiJ4JLE91T0maka5Smlvqy8zMWqCRw0r7ATekq0uHA1dHxE8krQGulzQPeAg4\nLbW/GTgR6ACeBc4EiIhtkr4ErEntzo+IbWn6LGAxMAK4JT3MzKxF6oZDRGwEDqlR/nvg2BrlAZzd\nTV+LgEU1ytuBgxsYr5mZDQB/Q9rMzDIOBzMzyzgczMws43AwM7OMw8HMzDIOBzMzyzgczMws43Aw\nM7OMw8HMzDIOBzMzyzgczMws43AwM7OMw8HMzDIOBzMzyzgczMws43AwM7OMw8HMzDIOBzMzyzgc\nzMws43AwM7OMw8HMzDIOBzMzyzgczMws03A4SBom6U5JN6X5AyStlrRB0nWSdk3lu6X5jlQ/udTH\nOan8fknHl8pnpbIOSQv67+WZmVlvNLPn8HFgfWn+YuAbETEFeByYl8rnAY9HxEHAN1I7JE0F5gDv\nAGYB30mBMwz4NnACMBU4PbU1M7MWaSgcJE0E3gdcluYFHAMsTU2WAKek6dlpnlR/bGo/G7g2Il6I\niAeADmB6enRExMaIeBG4NrU1M7MWaXTP4R+AzwCvpPk3A09ExPY03wlMSNMTgM0Aqf7J1P7V8qpl\nuivPSJovqV1Se1dXV4NDNzOzZtUNB0knAVsj4o5ycY2mUaeu2fK8MGJhREyLiGnjxo3rYdRmZtYX\nwxtoczRwsqQTgd2BkRR7EqMkDU97BxOBLal9JzAJ6JQ0HNgH2FYqrygv0125mZm1QN09h4g4JyIm\nRsRkihPKt0bEGcDPgFNTszbgxjS9LM2T6m+NiEjlc9LVTAcAU4DbgTXAlHT1065pHcv65dWZmVmv\nNLLn0J3PAtdKugC4E7g8lV8OXCmpg2KPYQ5ARKyTdD1wH7AdODsiXgaQ9FFgOTAMWBQR6/owLjMz\n66OmwiEiVgGr0vRGiiuNqts8D5zWzfIXAhfWKL8ZuLmZsZiZ2Y7jb0ibmVnG4WBmZhmHg5mZZRwO\nZmaWcTiYmVnG4WBmZhmHg5mZZRwOZmaWcTiYmVnG4WBmZhmHg5mZZRwOZmaWcTiYmVnG4WBmZhmH\ng5mZZRwOZmaWcTiYmVnG4WBmZhmHg5mZZRwOZmaWcTiYmVnG4WBmZpm64SBpd0m3S7pb0jpJX0zl\nB0haLWmDpOsk7ZrKd0vzHal+cqmvc1L5/ZKOL5XPSmUdkhb0/8s0M7NmNLLn8AJwTEQcAhwKzJI0\nA7gY+EZETAEeB+al9vOAxyPiIOAbqR2SpgJzgHcAs4DvSBomaRjwbeAEYCpwemprZmYtUjccovB0\nmt0lPQI4BliaypcAp6Tp2WmeVH+sJKXyayPihYh4AOgApqdHR0RsjIgXgWtTWzMza5GGzjmkT/h3\nAVuBFcBvgSciYntq0glMSNMTgM0Aqf5J4M3l8qpluiuvNY75ktoltXd1dTUydDMz64WGwiEiXo6I\nQ4GJFJ/0316rWXpWN3XNltcax8KImBYR08aNG1d/4GZm1itNXa0UEU8Aq4AZwChJw1PVRGBLmu4E\nJgGk+n2AbeXyqmW6KzczsxZp5GqlcZJGpekRwHHAeuBnwKmpWRtwY5peluZJ9bdGRKTyOelqpgOA\nKcDtwBpgSrr6aVeKk9bL+uPFmZlZ7wyv34TxwJJ0VdGbgOsj4iZJ9wHXSroAuBO4PLW/HLhSUgfF\nHsMcgIhYJ+l64D5gO3B2RLwMIOmjwHJgGLAoItb12ys0M7Om1Q2HiLgHOKxG+UaK8w/V5c8Dp3XT\n14XAhTXKbwZubmC8ZmY2APwNaTMzyzgczMws43AwM7OMw8HMzDIOBzMzyzgczMws43AwM7OMw8HM\nzDIOBzMzyzgczMws43AwM7OMw8HMzDIOBzMzyzgczMws43AwM7OMw8HMzDIOBzMzyzgczMws43Aw\nM7OMw8HMzDIOBzMzyzgczMws43AwM7NM3XCQNEnSzyStl7RO0sdT+RhJKyRtSM+jU7kkXSKpQ9I9\nkg4v9dWW2m+Q1FYqP0LS2rTMJZK0I16smZk1ppE9h+3ApyLi7cAM4GxJU4EFwMqImAKsTPMAJwBT\n0mM+cCkUYQKcBxwFTAfOqwRKajO/tNysvr80MzPrrbrhEBEPR8Sv0vRTwHpgAjAbWJKaLQFOSdOz\ngSuicBswStJ44HhgRURsi4jHgRXArFQ3MiJ+GREBXFHqy8zMWqCpcw6SJgOHAauB/SLiYSgCBNg3\nNZsAbC4t1pnKeirvrFFea/3zJbVLau/q6mpm6GZm1oSGw0HSXsA/An8TEX/oqWmNsuhFeV4YsTAi\npkXEtHHjxtUbspmZ9VJD4SBpF4pguCoifpSKH02HhEjPW1N5JzCptPhEYEud8ok1ys3MrEUauVpJ\nwOXA+oj4eqlqGVC54qgNuLFUPjddtTQDeDIddloOzJQ0Op2IngksT3VPSZqR1jW31JeZmbXA8Aba\nHA18CFgr6a5U9rfARcD1kuYBDwGnpbqbgROBDuBZ4EyAiNgm6UvAmtTu/IjYlqbPAhYDI4Bb0sPM\nzFqkbjhExC+ofV4A4Nga7QM4u5u+FgGLapS3AwfXG4uZmQ0Mf0PazMwyDgczM8s4HMzMLONwMDOz\njMPBzMwyDgczM8s4HMzMLONwMDOzjMPBzMwyDgczM8s4HMzMLONwMDOzjMPBzMwyDgczM8s4HMzM\nLONwMDOzjMPBzMwyDgczM8s4HMzMLONwMDOzjMPBzMwyDgczM8vUDQdJiyRtlXRvqWyMpBWSNqTn\n0alcki6R1CHpHkmHl5ZpS+03SGorlR8haW1a5hJJ6u8XaWZmzWlkz2ExMKuqbAGwMiKmACvTPMAJ\nwJT0mA9cCkWYAOcBRwHTgfMqgZLazC8tV70uMzMbYHXDISJ+DmyrKp4NLEnTS4BTSuVXROE2YJSk\n8cDxwIqI2BYRjwMrgFmpbmRE/DIiArii1JeZmbVIb8857BcRDwOk531T+QRgc6ldZyrrqbyzRnlN\nkuZLapfU3tXV1cuhm5lZPf19QrrW+YLoRXlNEbEwIqZFxLRx48b1cohmZlZPb8Ph0XRIiPS8NZV3\nApNK7SYCW+qUT6xRbmZmLdTbcFgGVK44agNuLJXPTVctzQCeTIedlgMzJY1OJ6JnAstT3VOSZqSr\nlOaW+jIzsxYZXq+BpGuA9wBjJXVSXHV0EXC9pHnAQ8BpqfnNwIlAB/AscCZARGyT9CVgTWp3fkRU\nTnKfRXFF1AjglvQwM7MWqhsOEXF6N1XH1mgbwNnd9LMIWFSjvB04uN44zMxs4Pgb0mZmlnE4mJlZ\nxuFgZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmYZh4OZmWUcDmZm\nlnE4mJlZxuFgZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmYZh4OZmWUcDmZmlnE4mJlZxuFgZmaZQRMO\nkmZJul9Sh6QFrR6PmdlQNrzVAwCQNAz4NvBfgU5gjaRlEXFfa0fWvyYv+OeWrXvTRe9r2brNbOcz\nWPYcpgMdEbExIl4ErgVmt3hMZmZD1qDYcwAmAJtL853AUdWNJM0H5qfZpyXd38v1jQUe6+WyA6nf\nxqmL+6OXHu0s2xQ81h3FY+1//T3OtzbacLCEg2qURVYQsRBY2OeVSe0RMa2v/exoO8s4wWPdUTzW\nHWNnGWsrxzlYDit1ApNK8xOBLS0ai5nZkDdYwmENMEXSAZJ2BeYAy1o8JjOzIWtQHFaKiO2SPgos\nB4YBiyJi3Q5cZZ8PTQ2QnWWc4LHuKB7rjrGzjLVl41REdmjfzMyGuMFyWMnMzAYRh4OZmWWGVDgM\ntlt0SJok6WeS1ktaJ+njqXyMpBWSNqTn0alcki5J479H0uEDPN5hku6UdFOaP0DS6jTO69LFBEja\nLc13pPrJAzzOUZKWSvp12rbvHMTb9BPpZ3+vpGsk7T5YtqukRZK2Srq3VNb0dpTUltpvkNQ2gGP9\nWvoduEfSDZJGlerOSWO9X9LxpfId/h5Ra6yluk9LCklj03zrtmtEDIkHxYnu3wIHArsCdwNTWzym\n8cDhaXpv4DfAVOCrwIJUvgC4OE2fCNxC8b2QGcDqAR7vJ4GrgZvS/PXAnDT9XeCsNP3XwHfT9Bzg\nugEe5xLgI2l6V2DUYNymFF/+fAAYUdqeHx4s2xV4F3A4cG+prKntCIwBNqbn0Wl69ACNdSYwPE1f\nXBrr1PT3vxtwQHpfGDZQ7xG1xprKJ1FclPMgMLbV23VA/ggGwwN4J7C8NH8OcE6rx1U1xhsp7i91\nPzA+lY0H7k/T3wNOL7V/td0AjG0isBI4Brgp/bI+Vvrje3X7pl/wd6bp4amdBmicI9MbrqrKB+M2\nrdwZYEzaTjcBxw+m7QpMrnrDbWo7AqcD3yuVv67djhxrVd0HgKvS9Ov+9ivbdSDfI2qNFVgKHAJs\n4rVwaNl2HUqHlWrdomNCi8aSSYcIDgNWA/tFxMMA6Xnf1KyVr+EfgM8Ar6T5NwNPRMT2GmN5dZyp\n/snUfiAcCHQB30+HwC6TtCeDcJtGxO+A/wM8BDxMsZ3uYHBu14pmt+Ng+bv7K4pP4DAIxyrpZOB3\nEXF3VVXLxjqUwqGhW3S0gqS9gH8E/iYi/tBT0xplO/w1SDoJ2BoRdzQ4llZu6+EUu+yXRsRhwDMU\nhz+607KxpuP1sykObewP7Amc0MN4Bu3vMN2PreVjlnQusB24qlJUo1nLxippD+Bc4PO1qmuUDchY\nh1I4DMpbdEjahSIYroqIH6XiRyWNT/Xjga2pvFWv4WjgZEmbKO6YewzFnsQoSZUvUpbH8uo4U/0+\nwLYBGGdl3Z0RsTrNL6UIi8G2TQGOAx6IiK6IeAn4EfCfGZzbtaLZ7djSv7t0ovYk4IxIx196GFOr\nxvqfKD4g3J3+xiYCv5L0llaOdSiFw6C7RYckAZcD6yPi66WqZUDl6oM2inMRlfK56QqGGcCTlV38\nHSkizomIiRExmWK73RoRZwA/A07tZpyV8Z+a2g/Ip8WIeATYLOltqehY4D4G2TZNHgJmSNoj/S5U\nxjrotmtJs9txOTBT0ui0pzQzle1wkmYBnwVOjohnq17DnHT11wHAFOB2WvQeERFrI2LfiJic/sY6\nKS5UeYRWbtcdcbJlsD4ozvz/huKKhHMHwXj+nGJX8B7grvQ4keI48kpgQ3oek9qL4p8i/RZYC0xr\nwZjfw2tXKx1I8UfVAfwQ2C2V757mO1L9gQM8xkOB9rRd/4niao5BuU2BLwK/Bu4FrqS4gmZQbFfg\nGopzIS9RvGHN6812pDje35EeZw7gWDsojstX/ra+W2p/bhrr/cAJpfId/h5Ra6xV9Zt47YR0y7ar\nb59hZmaZoXRYyczMGuRwMDOzjMPBzMwyDgczM8s4HMzMLONwsDckSS9LukvFHU/vlvRJSf32+y7p\nw5L2L81fJmlqP/V9iqRa35btaZmfVu6QatYffCmrvSFJejoi9krT+1LcTfbfI+K8JvoYFhEvd1O3\nCvh0RLT3x3ir+v4Pii9uPdbEMm3AxIi4sL/HY0OT9xzsDS8itgLzgY+mb5p+WNK3KvWSbpL0njT9\ntKTzJa0G3inp85LWqPh/CwvT8qcC04Cr0t7JCEmrJE1LfZwuaW1a5uLSep6WdGHak7lN0n7VY5X0\nx8ALlWCQtFjSpSr+78dGSe9W8f8A1ktaXFp0GcWdOs36hcPBhoSI2Ejx+75vnaZ7UtxK+aiI+AXw\nrYg4MiIOBkYAJ0XEUopvYJ8REYdGxHOVhdOhposp7j91KHCkpFNKfd8WEYcAPwf+R431Hw38qqps\ndOrvE8CPgW8A7wD+VNKh6fU9DuwmaaDv0mpvUA4HG0pq3cmy2ssUN0KseK+K/7q2luIN+h11lj8S\nWBXFzfQqdwJ9V6p7keJ/NkBxa+7JNZYfT3HL8bIfR3H8dy3waBT34nkFWFfVx1aKu7ua9dnw+k3M\ndn6SDqR4499Kcfvm8gej3UvTz1fOM0jaHfgOxf1sNkv6QlXbmqvqoe6leO0k38vU/vt7juJuq2Uv\npOdXStOV+XIfu6flzfrMew72hidpHMW/2/xWenPeBBwq6U2SJgHTu1m0EgSPqfifG6eW6p6i+Neu\n1VYD75Y0VtIwivMA/9rEcNcDBzXRHnj1Dr9voXhtZn3mPQd7oxoh6S5gF4o9hSuBym3R/53iX4mu\npbgbavUxfgAi4glJ/y+120RxS+eKxcB3JT1H8e8lK8s8LOkcittuC7g5Im6kcT8H/l6SSnsZjTiC\n4nzG9rotzRrgS1nNBhlJ36Q4z/DTJpdZFhErd9zIbCjxYSWzwefLwB5NLnOvg8H6k/cczMws4z0H\nMzPLOBzMzCzjcDAzs4zDwczMMg4HMzPL/H8+ldwCCZFHYQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107ff7c88>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## 使用本框及新框来收集所有骑行时长并制成列表。##\n",
    "## 使用 pyplot 函数来为骑行时长生成直方图。 ##\n",
    "%matplotlib inline \n",
    "\n",
    "# 直方图示例，数据来自湾区样本\n",
    "data = []\n",
    "with open('./data/Washington-2016-Summary.csv') as file:\n",
    "    reader = csv.DictReader(file)\n",
    "    for row in reader:\n",
    "        data.append(float(row['duration']))\n",
    "print(len(data))\n",
    "plt.hist(data)\n",
    "plt.title('Distribution of Trip Durations')\n",
    "plt.xlabel('Duration (m)')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "如果你按示例的用法来用 `.hist()` 和 `.show()` 函数，就可能得到一张完全在意料之外的统计图，该图左侧将有一条极长的柱形，相邻的第二条柱形则可能十分短小，而且中间和右边将空出许多空间，这个时候你可以看看 x 轴的时长值，你会发现，数据里有些高度异常的极端值，不过你不需要重新处理数据，只需为 `.hist()` 函数再加一些参数，限制绘图数据的范围即可，函数相关文件请见[[这里]](https://matplotlib.org/devdocs/api/_as_gen/matplotlib.pyplot.hist.html#matplotlib.pyplot.hist).\n",
    "\n",
    "**问题 5**：用 `.hist()` 函数的参数来为所选城市的会员骑行时长绘制分布图，并单独为散客做同样的操作。限制统计图的范围，仅绘制骑行时长小于 75 分钟的数据。另外，你可以设置每条柱形之间有五分钟的间隔。在每组数据中，各分布图的峰值分别在哪？你会如何描述每个分布图的形状？\n",
    "\n",
    "**答案**：华盛顿会员的时长峰值在5分钟，值为17500；散客的时长峰值在15分钟，值为2000。二者都呈右偏态分布"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYcAAAEWCAYAAACNJFuYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzt3Xu8FXW9//HXO1C0REXZGgIKGnq8\nVJhk9vBUlqZ4Sa2fnSBL8tiPLO3X9RRmJ820rI518lT2IyM0b5kektQy5GR28cJWUSA1tki6hWAj\nXtNI8HP++H6XDnvWvq4Fa8F+Px+P9Vgz35n5zmfNunzm+51ZM4oIzMzMil7R6ADMzKz5ODmYmVmJ\nk4OZmZU4OZiZWYmTg5mZlTg5mJlZiZNDE5H0A0n/Xqe6dpX0rKRBefwWSR+uR925vl9KmlKv+vqw\n3nMlrZL01zrV96Ckt9SjrkbZGK9B0laSHpC0Ux3qOlfSzDqE1d06BuXP/655/DJJZ9ex/gvr+X1q\nRk4OG4mkpZKel/SMpCcl/VHSqZJeeg8i4tSI+Eov6zqsu3ki4pGI2CYi1tUh9rMlXdap/iMj4pJa\n6+5jHKOBzwD7RMSrO007Mf8YPJu384uF8We7qjMi9oqI3/UjlsGSQtLf8jpWSbpZ0nv7/sr6tN7S\nj1x/X0MffRS4OSJW5jh2lTQrv+6nJC2Q9MENHEOvRcS6/Pl/ZAOt4hvAlyQN3kD1N5yTw8b1rogY\nCuwGnA98HvhRvVeyGX9gdwMer/xAFUXE5fnHYBvgSGBZZTyXraeO22jfXP8/AZcBF0k6sz8VNfn7\n9hHgJ4Xxy4ElwK7AjsAUoPS+NMKG3o6SBkVEO/AQcMyGXFdDRYQfG+EBLAUO61R2IPAisF8enwmc\nm4eHA9cDTwKrgd+RkvlP8jLPA88CnwPGAAGcAjwC3FooG5zruwX4GnAn8BRwHbBDnnYI0F4tXmAi\n8A/ghby+ewv1fTgPvwL4IvAX0g/EpcB2eVoljik5tlXAmd1sp+3y8h25vi/m+g/Lr/nFHMfMbuoo\nvZ5c3g78G7AA+Eeh7JA8fC7wU+BnwDNAK/DaLtYxOL+uMZ3KJ+U4t+9cf2EdM/Pwa3IdJ+dt8z/5\ntV4D/DW/97cAe+f5P5bfh3/kbTCrymvYCrgQWA48BnwL2DJPOyy/r5/L23cZcFIhtmOA+/Nrbwc+\nlct3B/4GDCrM+3fy57bKtjkMWFpl2/dqOwNfyLE9DTxQWG4w8O+kH+Wn83K7FN6LjwFt+bHe+0NK\n3N8D5uZ1/gYYXVjnPsDNpO/aA8D/KUyrLPurvB0q8ZwF/LDRvy0b6uGWQwNFxJ2kL021/uLP5Gkt\nwM6kL0xExAdJPyTvirRX/I3CMm8D9gaO6GKVJwH/SvpCrSX9iPQU46+ArwI/zet7fZXZPpQfbyf9\nkGwDfLfTPP8M7AUcSmqO793FKv+LlCB2z6/nJODkiLiZ9VsEH+op9i5MyvVs18X09wBXADuQfqRn\n9XFP9OfAEOCNfVjmraSWx9F5/HpgHPBqYCF5jz0ivk/6Uf1q3gbvrlLXl4AJwOuA/YGDgTMK00cB\nW5M+A6eSWjrb5mk/Bk6J1Lp9HfDbXP5aoC3W76K8PS/7vtzd11dVt7OkfUmtlDdExLak96rSNfRv\nwAmkHZbtgQ+TklTFsaTt/tou1vkB0vYZDvyJvF0lDQXmkHZKdgJOBKZL2quw7PuBLwNDgdty2f1A\nte/DZsHJofGWkb4gnb0AjAB2i4gXIuJ3kXdXunF2RPwtIp7vYvpPImJhRPyNtAf2L5UD1jU6EfhW\nRCyJiGdJP0aTOv2ofjkino+Ie4F7qfKlyrG8DzgjIp6JiKXABUA9+7K/ExHt3WyjOyJiVkS8AHwT\n2JY+/NBHxN9Je5/V3tOunBURz+Xt82JEzMyv/+/A2cABkl7Vy7pOJH0OOiJ1v53D+tvv76TW6QsR\nMRtYA+yZp70A7CNpaESsjoi7c/n2pL3toveQfiTPAv4i6W5JB/ThNXe1ndeSWj/7ShocEQ9HxJK8\nzIeBL0TE4ryd5kfE6kKdX42IJ7p5b38REX+IiDWkna23ShpBSip/johLI2JtRNxFSvInFJadFRG3\n5fWuyWXP5G2zWXJyaLyRpB+Tzr5Jah7/WtISSdN6UdejfZj+F2AL0l5UrXbJ9RXrHkxq8VQUzy56\njtS66Gw4sGWVukbWIcaKXm+jvKf8GOn19YqkrUiJodp72uM681k238jv+dOkzwD0/n0aQffbb1Wn\nFkDxvXg36YfykXx225ty+ROkPeaX5OTxuYjYh/Q+LwJm9TJG6GI7R8SDpFbzOcBKSVdKqpx8MJrU\npdRjnb1Y51Ok7tVdSMeyDs4nijwp6UnSTsqIHuoeSur62yw5OTSQpDeSvri/7zwt7zl+JiJ2B94F\nfFrSoZXJXVTZU8ui2PzflbSnuIrUj/rKQlyDSN1Zva13GekLVqx7LbCih+U6W5Vj6lzXY32spzu9\n3kb5TLKRpNfXW8eT9sbn5fH1ti2pq2j9gNZvEZ4EHAW8g9T19ZpKOJXZe1j/cvq5/SLijog4ltS1\ncj1wVZ50H7BHV63MiOggtfBGS9qO8udpMOmgdVGX2zkiLouIg4GxwCDSsTJIP9B7dPcSeniJxXVu\nR9q+y3K9cyNi+8Jjm4g4vYe69ya1gjdLTg4NIGlbSceQvnyXRcSCKvMcI+k1kkQ6+LYuPyD96O7e\nj1V/QNI+kl5J2jO7Ju+1/RnYStLRkrYgHQQeUlhuBTCmeNptJ1cCn5I0VtI2vHyMYm1fgsuxXA2c\nJ2mopN2AT5MOCG4sB0o6Lm+Hz5K6Dub1sAySdsyncv4X8LWIqOxRzid3sUk6kNQd052hpOTyOOkH\n9rxO03t6768kHdMZLqmF1H3Y4/aTtLWk90vaNnf1PEP+vOXuvUeAAwrzf0PSvrmlsy3pVNcH8h75\nA8BQSUfk7XgWqZVaVHU7S9pb0tslDSEd2H+elz/3FwPnStpDyXhJfem+e5ekN+e6zwV+HxHLgdmk\nbqz3S9oiPw7sdMyhmrcBv+zD+jcpTg4b1y8kPUPaUzmTdCbJyV3MO4509sSzpL7d70fELXna14Av\n5ibwZ/uw/p+Qzoj6K6lf9//BS03sj5G+fI+R9vzaC8v9LD8/Luluymbkum8FHib1a3+8D3EVfTyv\nfwmpRXVFrn9jmUU6cLma1LXwnh6S3KL8P4rFpPfy4xFxTmH6maSDzU+Sfqiv6GH9PybtzS4jddX8\nsdP0i4HXS3pC0jVVlv8yaW92AWmP/w5e3vPuyRTS8YOnSWe+FY9V/P9O49uQznh7itTVswup1URE\nPEF6Hy8hfZ5Ws363InS9nYeQ/kOwKi8zjLSzAqmr9eekM46eBqaTPse9dRkpKawiHXD/YI73KdJJ\nHB8gtbz+StpmQ6pXA5JGkr6jv+jD+jcp6vkYp9nAIOlcYFQNZ0JttvKxlHuAt0WV/5kMNJK+AyyK\niOmNjmVDaeY/3ZhZk8hnTnV1+vGAExGfaHQMG5q7lczMrMTdSmZmVuKWg5mZlWyyxxyGDx8eY8aM\naXQYZmablLvuumtVRLT0NN8mmxzGjBlDa2tro8MwM9ukSPpLz3O5W8nMzKpwcjAzsxInBzMzK3Fy\nMDOzEicHMzMrcXIwM7MSJwczMytxcjAzs5Iek4OkGZJWSlpYKPuppPn5sVTS/Fw+RtLzhWk/KCxz\ngKQFktokXZhvYoOkHSTNkbQ4Pw/bEC/UzMx6rzf/kJ4JfBe4tFIQEe+rDEu6gHTDj4qHImJ8lXou\nAqYCtwM3AhNJd1GaRrpF3/n5PsnTgM/37WU01phpN9S9zqXnH133Os3MeqvHlkNE3EoXN0vPe///\nQro1YZckjQC2jYjb8v1yLyXfNQo4jnTHKPLz8VWqMDOzjajWYw5vAVZExOJC2VhJ90j6raS35LKR\nrH/byfZcBrBzvo8r+XmnrlYmaaqkVkmtHR0dNYZuZmZdqTU5TGb9VsNyYNeI2J90Y/gr8s3HVWXZ\nPt9IIiKmR8SEiJjQ0tLjRQXNzKyf+n1VVkmDgfcAB1TKImINsCYP3yXpIWBPUkthVGHxUaQbqAOs\nkDQiIpbn7qcBf39aM7NGq6XlcBjwQES81F0kqUXSoDy8OzAOWJK7i56RdFA+TnEScF1ebDYwJQ9P\nKZSbmVmD9OZU1iuB24C9JLVLOiVPmkT5QPRbgfsk3QtcA5waEZWD2R8FLgbagIdIZyoBnA+8U9Ji\n4J153MzMGqjHbqWImNxF+YeqlF0LXNvF/K3AflXKHwcO7SkOMzPbePwPaTMzK3FyMDOzEicHMzMr\ncXIwM7MSJwczMytxcjAzsxInBzMzK3FyMDOzEicHMzMrcXIwM7MSJwczMytxcjAzsxInBzMzK3Fy\nMDOzEicHMzMrcXIwM7MSJwczMytxcjAzsxInBzMzK3FyMDOzkh6Tg6QZklZKWlgoO1vSY5Lm58dR\nhWlnSGqT9KCkIwrlE3NZm6RphfKxku6QtFjSTyVtWc8XaGZmfdeblsNMYGKV8m9HxPj8uBFA0j7A\nJGDfvMz3JQ2SNAj4HnAksA8wOc8L8PVc1zjgCeCUWl6QmZnVrsfkEBG3Aqt7Wd9xwFURsSYiHgba\ngAPzoy0ilkTEP4CrgOMkCXgHcE1e/hLg+D6+BjMzq7NajjmcLum+3O00LJeNBB4tzNOey7oq3xF4\nMiLWdiqvStJUSa2SWjs6OmoI3czMutPf5HARsAcwHlgOXJDLVWXe6Ed5VRExPSImRMSElpaWvkVs\nZma9Nrg/C0XEisqwpB8C1+fRdmB0YdZRwLI8XK18FbC9pMG59VCc38zMGqRfLQdJIwqj7wYqZzLN\nBiZJGiJpLDAOuBOYB4zLZyZtSTpoPTsiAvgNcEJefgpwXX9iMjOz+umx5SDpSuAQYLikduAs4BBJ\n40ldQEuBjwBExCJJVwN/AtYCp0XEulzP6cBNwCBgRkQsyqv4PHCVpHOBe4Af1e3VmZlZv/SYHCJi\ncpXiLn/AI+I84Lwq5TcCN1YpX0I6m8nMzJqE/yFtZmYlTg5mZlbi5GBmZiVODmZmVuLkYGZmJU4O\nZmZW4uRgZmYlTg5mZlbi5GBmZiVODmZmVuLkYGZmJU4OZmZW4uRgZmYlTg5mZlbi5GBmZiVODmZm\nVuLkYGZmJU4OZmZW4uRgZmYlTg5mZlbSY3KQNEPSSkkLC2XflPSApPskzZK0fS4fI+l5SfPz4weF\nZQ6QtEBSm6QLJSmX7yBpjqTF+XnYhnihZmbWe71pOcwEJnYqmwPsFxGvA/4MnFGY9lBEjM+PUwvl\nFwFTgXH5UalzGjA3IsYBc/O4mZk1UI/JISJuBVZ3Kvt1RKzNo7cDo7qrQ9IIYNuIuC0iArgUOD5P\nPg64JA9fUig3M7MGqccxh38FflkYHyvpHkm/lfSWXDYSaC/M057LAHaOiOUA+XmnrlYkaaqkVkmt\nHR0ddQjdzMyqqSk5SDoTWAtcnouWA7tGxP7Ap4ErJG0LqMri0df1RcT0iJgQERNaWlr6G7aZmfVg\ncH8XlDQFOAY4NHcVERFrgDV5+C5JDwF7kloKxa6nUcCyPLxC0oiIWJ67n1b2NyYzM6uPfrUcJE0E\nPg8cGxHPFcpbJA3Kw7uTDjwvyd1Fz0g6KJ+ldBJwXV5sNjAlD08plJuZWYP02HKQdCVwCDBcUjtw\nFunspCHAnHxG6u35zKS3AudIWgusA06NiMrB7I+SznzamnSMonKc4nzgakmnAI8A763LKzMzs37r\nMTlExOQqxT/qYt5rgWu7mNYK7Fel/HHg0J7iMDOzjcf/kDYzsxInBzMzK3FyMDOzEicHMzMrcXIw\nM7OSfv8JzjasMdNuqGt9S88/uq71mdnmzS0HMzMrcXIwM7MSJwczMytxcjAzsxInBzMzK3FyMDOz\nEicHMzMrcXIwM7MSJwczMytxcjAzsxInBzMzK3FyMDOzEicHMzMrcXIwM7MSJwczMyvpVXKQNEPS\nSkkLC2U7SJojaXF+HpbLJelCSW2S7pP0hsIyU/L8iyVNKZQfIGlBXuZCSarnizQzs77pbcthJjCx\nU9k0YG5EjAPm5nGAI4Fx+TEVuAhSMgHOAt4EHAicVUkoeZ6pheU6r8vMzDaiXiWHiLgVWN2p+Djg\nkjx8CXB8ofzSSG4Htpc0AjgCmBMRqyPiCWAOMDFP2zYibouIAC4t1GVmZg1QyzGHnSNiOUB+3imX\njwQeLczXnsu6K2+vUl4iaaqkVkmtHR0dNYRuZmbd2RAHpKsdL4h+lJcLI6ZHxISImNDS0lJDiGZm\n1p1aksOK3CVEfl6Zy9uB0YX5RgHLeigfVaXczMwapJbkMBuonHE0BbiuUH5SPmvpIOCp3O10E3C4\npGH5QPThwE152jOSDspnKZ1UqMvMzBpgcG9mknQlcAgwXFI76ayj84GrJZ0CPAK8N89+I3AU0AY8\nB5wMEBGrJX0FmJfnOyciKge5P0o6I2pr4Jf5YWZmDdKr5BARk7uYdGiVeQM4rYt6ZgAzqpS3Avv1\nJhYzM9vw/A9pMzMrcXIwM7MSJwczMytxcjAzsxInBzMzK+nV2UqbmzHTbmh0CGZmTc0tBzMzK3Fy\nMDOzEicHMzMrcXIwM7MSJwczMytxcjAzsxInBzMzK3FyMDOzEicHMzMrcXIwM7MSJwczMytxcjAz\nsxInBzMzK3FyMDOzkn4nB0l7SZpfeDwt6ZOSzpb0WKH8qMIyZ0hqk/SgpCMK5RNzWZukabW+KDMz\nq02/7+cQEQ8C4wEkDQIeA2YBJwPfjoj/KM4vaR9gErAvsAtws6Q98+TvAe8E2oF5kmZHxJ/6G5uZ\nmdWmXjf7ORR4KCL+IqmreY4DroqINcDDktqAA/O0tohYAiDpqjyvk4OZWYPU65jDJODKwvjpku6T\nNEPSsFw2Eni0ME97LuuqvETSVEmtklo7OjrqFLqZmXVWc3KQtCVwLPCzXHQRsAepy2k5cEFl1iqL\nRzfl5cKI6RExISImtLS01BS3mZl1rR7dSkcCd0fECoDKM4CkHwLX59F2YHRhuVHAsjzcVbmZmTVA\nPbqVJlPoUpI0ojDt3cDCPDwbmCRpiKSxwDjgTmAeME7S2NwKmZTnNTOzBqmp5SDplaSzjD5SKP6G\npPGkrqGllWkRsUjS1aQDzWuB0yJiXa7ndOAmYBAwIyIW1RKXmZnVpqbkEBHPATt2KvtgN/OfB5xX\npfxG4MZaYjEzs/rxP6TNzKzEycHMzEqcHMzMrMTJwczMSup1+QxrcmOm3VDX+paef3Rd6zOz5uKW\ng5mZlTg5mJlZiZODmZmVODmYmVmJk4OZmZU4OZiZWYmTg5mZlTg5mJlZiZODmZmVODmYmVmJk4OZ\nmZU4OZiZWYmTg5mZlTg5mJlZiZODmZmV1JwcJC2VtEDSfEmtuWwHSXMkLc7Pw3K5JF0oqU3SfZLe\nUKhnSp5/saQptcZlZmb9V6+Ww9sjYnxETMjj04C5ETEOmJvHAY4ExuXHVOAiSMkEOAt4E3AgcFYl\noZiZ2ca3obqVjgMuycOXAMcXyi+N5HZge0kjgCOAORGxOiKeAOYAEzdQbGZm1oN6JIcAfi3pLklT\nc9nOEbEcID/vlMtHAo8Wlm3PZV2Vr0fSVEmtklo7OjrqELqZmVVTj3tIHxwRyyTtBMyR9EA386pK\nWXRTvn5BxHRgOsCECRNK083MrD5qbjlExLL8vBKYRTpmsCJ3F5GfV+bZ24HRhcVHAcu6KTczswao\nKTlIepWkoZVh4HBgITAbqJxxNAW4Lg/PBk7KZy0dBDyVu51uAg6XNCwfiD48l5mZWQPU2q20MzBL\nUqWuKyLiV5LmAVdLOgV4BHhvnv9G4CigDXgOOBkgIlZL+gowL893TkSsrjE2MzPrp5qSQ0QsAV5f\npfxx4NAq5QGc1kVdM4AZtcRjZmb14X9Im5lZiZODmZmVODmYmVmJk4OZmZU4OZiZWYmTg5mZlTg5\nmJlZiZODmZmVODmYmVmJk4OZmZXU45LdNgCNmXZDXetbev7Rda3PzGrjloOZmZU4OZiZWYmTg5mZ\nlTg5mJlZiZODmZmVODmYmVmJk4OZmZU4OZiZWYmTg5mZlfQ7OUgaLek3ku6XtEjSJ3L52ZIekzQ/\nP44qLHOGpDZJD0o6olA+MZe1SZpW20syM7Na1XL5jLXAZyLibklDgbskzcnTvh0R/1GcWdI+wCRg\nX2AX4GZJe+bJ3wPeCbQD8yTNjog/1RCbmZnVoN/JISKWA8vz8DOS7gdGdrPIccBVEbEGeFhSG3Bg\nntYWEUsAJF2V53VyMDNrkLocc5A0BtgfuCMXnS7pPkkzJA3LZSOBRwuLteeyrsrNzKxBak4OkrYB\nrgU+GRFPAxcBewDjSS2LCyqzVlk8uimvtq6pkloltXZ0dNQaupmZdaGm5CBpC1JiuDwi/hsgIlZE\nxLqIeBH4IS93HbUDowuLjwKWdVNeEhHTI2JCRExoaWmpJXQzM+tGLWcrCfgRcH9EfKtQPqIw27uB\nhXl4NjBJ0hBJY4FxwJ3APGCcpLGStiQdtJ7d37jMzKx2tZytdDDwQWCBpPm57AvAZEnjSV1DS4GP\nAETEIklXkw40rwVOi4h1AJJOB24CBgEzImJRDXGZmVmNajlb6fdUP15wYzfLnAecV6X8xu6WMzOz\njcv/kDYzsxInBzMzK3FyMDOzEicHMzMrqeVsJbO6GTPthrrXufT8o+tep9lA4ZaDmZmVODmYmVmJ\nk4OZmZU4OZiZWYmTg5mZlTg5mJlZiZODmZmVODmYmVmJk4OZmZU4OZiZWYmTg5mZlTg5mJlZiS+8\nZ5utel/Mzxfys4HELQczMytxcjAzsxInBzMzK2maYw6SJgLfAQYBF0fE+Q0OyWw9PoZhA0lTJAdJ\ng4DvAe8E2oF5kmZHxJ8aG5nZhuO731kza4rkABwItEXEEgBJVwHHAU4OZn2wIRLOQFPvBLuptjib\nJTmMBB4tjLcDb+o8k6SpwNQ8+qykB/u5vuHAqn4uu7E0e4zNHh80f4zNHh8MwBj19XrV9JJmi2+3\n3szULMlBVcqiVBAxHZhe88qk1oiYUGs9G1Kzx9js8UHzx9js8YFjrIdmj68rzXK2UjswujA+CljW\noFjMzAa8ZkkO84BxksZK2hKYBMxucExmZgNWU3QrRcRaSacDN5FOZZ0REYs24Cpr7praCJo9xmaP\nD5o/xmaPDxxjPTR7fFUpotS1b2ZmA1yzdCuZmVkTcXIwM7OSAZccJE2U9KCkNknTmiCeGZJWSlpY\nKNtB0hxJi/PzsAbHOFrSbyTdL2mRpE80U5yStpJ0p6R7c3xfzuVjJd2R4/tpPtmhoSQNknSPpOub\nLUZJSyUtkDRfUmsua4r3uBDj9pKukfRA/jy+uZlilLRX3n6Vx9OSPtlMMfbWgEoOhct0HAnsA0yW\ntE9jo2ImMLFT2TRgbkSMA+bm8UZaC3wmIvYGDgJOy9utWeJcA7wjIl4PjAcmSjoI+Drw7RzfE8Ap\nDYqv6BPA/YXxZovx7RExvnBefrO8xxXfAX4VEf8EvJ60LZsmxoh4MG+/8cABwHPArGaKsdciYsA8\ngDcDNxXGzwDOaIK4xgALC+MPAiPy8AjgwUbH2Cne60jXwWq6OIFXAneT/mG/Chhc7b1vUGyjSD8M\n7wCuJ/35s2liBJYCwzuVNc17DGwLPEw+kaYZY+wU1+HAH5o5xu4eA6rlQPXLdIxsUCzd2TkilgPk\n550aHM9LJI0B9gfuoInizN0184GVwBzgIeDJiFibZ2mG9/o/gc8BL+bxHWmuGAP4taS78qVqoIne\nY2B3oAP4ce6au1jSq5osxqJJwJV5uFlj7NJASw69ukyHVSdpG+Ba4JMR8XSj4ymKiHWRmvKjSBdy\n3LvabBs3qpdJOgZYGRF3FYurzNrIz+PBEfEGUrfraZLe2sBYqhkMvAG4KCL2B/5Gk3bP5GNHxwI/\na3Qs/TXQksOmcpmOFZJGAOTnlQ2OB0lbkBLD5RHx37m46eKMiCeBW0jHRraXVPmjZ6Pf64OBYyUt\nBa4idS39J00UY0Qsy88rSf3kB9Jc73E70B4Rd+Txa0jJoplirDgSuDsiVuTxZoyxWwMtOWwql+mY\nDUzJw1NIffwNI0nAj4D7I+JbhUlNEaekFknb5+GtgcNIByp/A5zQ6PgAIuKMiBgVEWNIn7v/iYgT\naZIYJb1K0tDKMKm/fCFN8h4DRMRfgUcl7ZWLDiVd1r9pYiyYzMtdStCcMXav0Qc9NvYDOAr4M6lP\n+swmiOdKYDnwAmnP6BRSX/RcYHF+3qHBMf4zqbvjPmB+fhzVLHECrwPuyfEtBL6Uy3cH7gTaSM37\nIY1+v3NchwDXN1OMOY5782NR5bvRLO9xIc7xQGt+r38ODGvCGF8JPA5sVyhrqhh78/DlM8zMrGSg\ndSuZmVkvODmYmVmJk4OZmZU4OZiZWYmTg5mZlTg52GZJ0rp8VcxF+Wqtn5ZUt8+7pA9J2qUwfnG9\nLuIo6XhJX+rjMjdvClf6tE2HT2W1zZKkZyNimzy8E3AF6SJoZ/WhjkERsa6LabcAn42I1nrE26nu\nPwLHRsSqPiwzBRgVEefVOx4bmNxysM1epMtBTAVOV/IhSd+tTJd0vaRD8vCzks6RdAfwZklfkjRP\n0kJJ0/PyJwATgMtz62RrSbdImpDrmJzvi7BQ0tcL63lW0nm5JXO7pJ07xyppT2BNJTFIminpIqX7\naSyR9Dale4DcL2lmYdHZpH/lmtWFk4MNCBGxhPR57+lqmK8iXT79TRHxe+C7EfHGiNgP2Bo4JiKu\nIf1L98RI1+5/vrJw7mr6OunaSeOBN0o6vlD37ZHuO3Er8H+rrP9g0iXHi4bl+j4F/AL4NrAv8FpJ\n4/PrewIYImnHXmwOsx45OdhAUu0qqJ2tI11gsOLtSndqW0D6gd63h+XfCNwSER2RLsV9OVC5uuk/\nSPdxALiLdB+PzkaQLktd9ItI/b8LgBURsSAiXiRd5qJYx0pgF8zqYHDPs5ht+iTtTvrhX0m6s11x\nx2irwvDfK8cZJG0FfB+YEBGPSjq707xVV9XNtBfi5YN866j+/Xse2K5T2Zr8/GJhuDJerGOrvLxZ\nzdxysM2epBbgB6QuoiDd8WyVC3kTAAAA4ElEQVS8pFdIGk26NHU1lUSwKt/L4oTCtGeAoVWWuQN4\nm6Th+ba0k4Hf9iHc+4HX9GF+4KUr576a9NrMauaWg22uts53htuC1FL4CVC53PgfSLebXEC6imvn\nPn4g3RtC0g/zfEtJl3yvmAn8QNLzpNt7VpZZLukM0qW4BdwYEX25PPOtwAWSVGhl9MYBpOMZa3uc\n06wXfCqrWZOR9B3ScYab+7jM7IiYu+Eis4HE3UpmzeerpHsC9MVCJwarJ7cczMysxC0HMzMrcXIw\nM7MSJwczMytxcjAzsxInBzMzK/lfbWAGV27WrAcAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x107ff7e10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYEAAAEWCAYAAACAOivfAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAH7JJREFUeJzt3XucHFWd9/HP1yAXBSSYAUMSDNHA\nErwEGRFfeEFB7gK6XpJlJbr4RFzwUcF1g7iCKIq6wCOPa9goMaBcRJAlIl5CVkRXCQwYciFGQoww\nJCaD4ZIABhJ++8c5LZVJz0zPdGe6M/V9v179mu5TVad+1dXTvzqnqusoIjAzs3J6QbMDMDOz5nES\nMDMrMScBM7MScxIwMysxJwEzsxJzEjAzKzEngSaSdJmkf2tQXXtLWi9pWH59m6QPN6LuXN9PJE1p\nVH39WO8XJT0i6c8Nqm+ppDc3oq5mGYxtkLSjpN9L2mNrrqfVSBop6T5J2zc7lsHiJLCVSFoh6WlJ\n6yQ9Juk3kk6T9Lf3PCJOi4gv1FjXEb3NExEPRsTOEbGpAbGfJ+l73eo/JiKuqLfufsYxBjgLmBAR\nL+s27eSc9Nbn9/m5wuv1PdUZEftFxK8GEMt2kkLSk3kdj0i6VdJ7+79l/Vrv9ySdVywb6Db000eB\nWyNiTSGWQyT9VNLjktZKmifplHpWIukISSvqDbZRImIV8Gvg1GbHMlicBLaud0bELsDLgQuBfwUu\nb/RKJG3X6DpbxMuBvxS/iCoi4qqc9HYGjgFWVl7nss008D06INf/d8D3gOmSzhlIRS2+3z4CfLfy\nQtKbgFuBucA44KXAGcCxTYluKyjsj6tI218OEeHHVngAK4AjupUdDDwHvCq/ngV8MT8fAdwMPAas\nBX5FStLfzcs8DawHPg2MBYJ0tPIgcHuhbLtc323Al4E7gceBm4Dd87TDgM5q8QJHA88Az+b13Vuo\n78P5+QuAzwJ/AtYAVwIvydMqcUzJsT0CnNPL+/SSvHxXru+zuf4j8jY/l+OY1UsdW2xPLu8E/gVY\nCDxTKDssP/8i8H3gB8A6oAN4dQ/r2C5v19hu5ZNynLt1r7+wjln5+StzHR/K781/5229Hvhz3ve3\nAfvn+f8574dn8ntwY5Vt2BG4FFgFPAxcDGyfpx2R9+un8/u7EjilENvxwJK87Z3AJ3P5OOBJYFhh\n3juAr/eyDz4M3NbT+1VtXXnfF/fxemCPGrfp7MI2vTPXfz/pf+fThTheAHwGeID0WbwWGN7T/sjl\n2wN/BUY1+3tkMB5uCQyiiLiT9A9QrT/3rDytDdiT9MGNiPgA6QP6zkhHuV8tLPNWYH/gqB5WeQrw\nT8BewEbSP1ZfMf4U+BLw/by+11aZ7YP58TbSF8bOwDe6zfMmYD/gcOBzkvbvYZX/n/RlMC5vzynA\nhyLiVjY/wv9gX7H3YFKu5yU9TH83cDWwO+nL+MZ+HqH/F7AD8Pp+LPMWUkviuPz6ZmA88DJgEfkI\nPCK+SUpSX8rvwbuq1PU5oB14DXAgcCjpC7JiNLAT6TNwGqnlsmue9h3g1Eit1dcAv8zlrwaWRe5a\nlLQL6QDm+n5sY3dbrCsiHid9gVe6MneO1OqrZZtekLfpC6TW9aQ872HA+ZL2zvOeSXqf35KXe5It\n/w822x8R8QywHKj22R9ynAQG30rSF053zwIjgZdHxLMR8avIhyW9OC8inoyIp3uY/t2IWBQRTwL/\nBryvcuK4TicDF0fE8ohYT/oHndTty/PzEfF0RNwL3EuVf6gcy/uBsyNiXUSsAC4CPtCAGCu+HhGd\nvbxH8yLixoh4FvgasCv9+EKPiL+Sjj6r7dOenBsRT+X357mImJW3/6/AecBBkl5cY10nkz4HXfkL\n9Hw2f//+SmptPhsRs4ENwL552rPABEm7RMTaiLgnl+9GOmKv2B0Q6ch8oHpa10C36cK8z64lHThd\nEhHrI2IBsJSUQCB163wmIh4uvL/vK56bo7A/CmXrSO/DkOckMPhGkb40uvsasAz4uaTlkqbVUNdD\n/Zj+J+CFpG6neu2V6yvWvR2pBVNRvJrnKVJrobsRpKZ397pGNSDGiprfo3zk+zBp+2oiaUfSl2S1\nfdrnOiUNk/TVvM+fIH0GoPb9NJLe379HYvOLBYr74l3ACcCD+WqyN+TyR4FdCsusJXWbjKwxpmp6\nWlc1/dmmyhf36sL0p3l+G/cGfpQvzniM1DUYpG6nimqfkV1I3XNDnpPAIJL0etKH+dfdp+UjwbMi\nYhypiXympMMrk3uosq+WwpjC871JR2OPkJrELyrENYx0NFVrvStJJ22LdW9k83/EWjySY+pe18P9\nrKc3Nb9H+ehwFGn7anUS6ej6rvx6s/eW1MWzeUCbt/BOIZ1cfTupy+qVlXAqs/ex/lUM8P2LiHkR\ncQLpC/Fm0lE1wALgFZVWY0SsI51b+vtequt1u3tZV7XtG/A2VdEJvCMidis8doyIvx2kdG9x58tD\nx5FasEOek8AgkLSrpONJH/zvRcTCKvMcL+mVkgQ8AWzKD0hfruMGsOp/lDRB0otITerr8xHUH4Ad\nJR0n6YWkk7E7FJZbDYzt1mQuugb4pKR9JO3M8+cQNvYnuBzLdcAFknaR9HJSH+73el+yoQ6WdGJ+\nHz5F6ga4q49lkPRSSR8gndP4ckRUjhrnk7vGJB1MOufQm11ISeQvpC/RC7pN72vfX0M65zJCUhup\n26/P90/STpL+QdKuuVtlHfnzlrvlHgQOKizyL8CHJZ0pafdcx4GSrs7T7wVeI+nVknYCzq1lXXn7\nRuTzDnVtUw8uA75UOUcgaQ9JJ/SxzCHAHyKikQcjLctJYOv6kaR1pObmOaSrHD7Uw7zjSZfgrQd+\nC3wzIm7L074MfDY3aT/Vj/V/l3QF0p9JV1z8X4B8Qu6fgW+TjrCeJB0xVfwg//2LpGp9tzNz3bcD\nfyT10X6sH3EVfSyvfzmphXR1rn+w3Aj8I6nL4/3Au/tIZovz7xDuJ+3Lj0XE+YXp55BOMj5G+vK6\nessqNvMdUstjJbAY+E236d8GXivpUUnVTsx+nvQFvJB0BD+P9HmpxRTgT7kb6lQ273f/z+LrSL9L\nOIJ0EcIKSWuB6cAtefp9pIOB20h98rfXsq6IWATckOt8TOnHafVsU3cXAz8F5ub/xd/Q9zmfk0nJ\noxTU97lHs6FJ0heB0XVceTRk5XMdvwPeGlV+pzFUSRpJ+i3ExHyV0JDXyj9WMbMmyVfS9HRZ75AV\n6RfDE5odx2Byd5CZWYm5O8jMrMTcEjAzK7GWPycwYsSIGDt2bLPDMDPbZtx9992PRERb33NuA0lg\n7NixdHR0NDsMM7NthqQ/9T1X4u4gM7MScxIwMysxJwEzsxJzEjAzKzEnATOzEuszCUgaI+kXkpZI\nWizp47l8d0lzJN2f/w7P5ZJ0qaRlkhZIel2hril5/vslTdl6m2VmZrWopSWwETgrIvYn3WL1dEkT\ngGnA3IgYT7rhUmUQlGNId8QcD0wl3WmQfPvZc4E3kIaqO7eSOMzMrDn6TAIRsaoyFFweXGIJaeCN\nE4Er8mxXkAbXIJdfGckdwG75znxHAXPy0HKPAnNIg5qbmVmT9OucgKSxpMGc5wF75jvuVe68Vxmu\nbRSbD9fWmct6Kq+2nqmSOiR1dHV19SdEMzPrh5p/MZxHkLoB+EREPJEGwKo+a5Wy6KV8y8KIGcAM\ngPb2dt/hrh/GTvtxQ+tbceFxDa3PzFpLTS2BPPTeDcBVEfHDXLw6d/NUBmKoDDzRyeZj244mjZrU\nU7mZmTVJLVcHCbgcWBIRFxcmzSYNGUf+e1Oh/JR8ldAhwOO5u+hnwJGShucTwkfmMjMza5JauoMO\nJY0HulDS/Fz2GeBC4DpJp5IGpX5vnnYLcCywDHiKPKZuRKyV9AWeH8T7/IhY25CtMDOzAekzCUTE\nr6nenw9weJX5Azi9h7pmMriDiJuZWS/8i2EzsxJzEjAzKzEnATOzEnMSMDMrMScBM7MSa/kxhq25\n/Atks6HNLQEzsxJzEjAzKzEnATOzEnMSMDMrMScBM7MScxIwMysxJwEzsxJzEjAzKzH/WKyJGv1D\nLDOz/nJLwMysxJwEzMxKrJYxhmdKWiNpUaHs+5Lm58eKyrCTksZKerow7bLCMgdJWihpmaRL89jF\nZmbWRLWcE5gFfAO4slIQEe+vPJd0EfB4Yf4HImJilXqmA1OBO0jjEB8N/KT/IZuZWaP02RKIiNuB\nqgPC56P59wHX9FaHpJHArhHx2zwG8ZXASf0P18zMGqnecwJvBlZHxP2Fsn0k/U7SLyW9OZeNAjoL\n83TmsqokTZXUIamjq6urzhDNzKwn9SaByWzeClgF7B0RBwJnAldL2hWo1v8fPVUaETMioj0i2tva\n2uoM0czMejLg3wlI2g54N3BQpSwiNgAb8vO7JT0A7Es68h9dWHw0sHKg6zYzs8aopyVwBPD7iPhb\nN4+kNknD8vNxwHhgeUSsAtZJOiSfRzgFuKmOdZuZWQPUconoNcBvgf0kdUo6NU+axJYnhN8CLJB0\nL3A9cFpEVE4qfxT4NrAMeABfGWRm1nR9dgdFxOQeyj9YpewG4IYe5u8AXtXP+MzMbCvyL4bNzErM\nScDMrMScBMzMSsxJwMysxJwEzMxKzEnAzKzEnATMzErMScDMrMScBMzMSsxJwMysxJwEzMxKzEnA\nzKzEnATMzErMScDMrMQGPLKY2UCMnfbjhte54sLjGl6nWVm4JWBmVmJOAmZmJVbL8JIzJa2RtKhQ\ndp6khyXNz49jC9POlrRM0lJJRxXKj85lyyRNa/ymmJlZf9XSEpgFHF2l/JKImJgftwBImkAae/iA\nvMw3JQ3Lg8//B3AMMAGYnOc1M7MmqmWM4dslja2xvhOBayNiA/BHScuAg/O0ZRGxHEDStXne+/od\nsZmZNUw95wTOkLQgdxcNz2WjgIcK83Tmsp7Kq5I0VVKHpI6urq46QjQzs94MNAlMB14BTARWARfl\nclWZN3opryoiZkREe0S0t7W1DTBEMzPry4B+JxARqyvPJX0LuDm/7ATGFGYdDazMz3sqNzOzJhlQ\nS0DSyMLLdwGVK4dmA5Mk7SBpH2A8cCdwFzBe0j6StiedPJ498LDNzKwR+mwJSLoGOAwYIakTOBc4\nTNJEUpfOCuAjABGxWNJ1pBO+G4HTI2JTrucM4GfAMGBmRCxu+NaYmVm/1HJ10OQqxZf3Mv8FwAVV\nym8BbulXdGZmtlX5F8NmZiXmJGBmVmJOAmZmJeYkYGZWYk4CZmYl5iRgZlZiTgJmZiXmJGBmVmJO\nAmZmJeYkYGZWYk4CZmYl5iRgZlZiTgJmZiXmJGBmVmJOAmZmJTag4SXLauy0Hzc7BDOzhnJLwMys\nxPpMApJmSlojaVGh7GuSfi9pgaQbJe2Wy8dKelrS/Py4rLDMQZIWSlom6VJJ2jqbZGZmtaqlJTAL\nOLpb2RzgVRHxGuAPwNmFaQ9ExMT8OK1QPh2YShp8fnyVOs3MbJD1mQQi4nZgbbeyn0fExvzyDmB0\nb3VIGgnsGhG/jYgArgROGljIZmbWKI04J/BPwE8Kr/eR9DtJv5T05lw2CugszNOZy6qSNFVSh6SO\nrq6uBoRoZmbV1JUEJJ0DbASuykWrgL0j4kDgTOBqSbsC1fr/o6d6I2JGRLRHRHtbW1s9IZqZWS8G\nfImopCnA8cDhuYuHiNgAbMjP75b0ALAv6ci/2GU0Glg50HWbmVljDKglIOlo4F+BEyLiqUJ5m6Rh\n+fk40gng5RGxClgn6ZB8VdApwE11R29mZnXpsyUg6RrgMGCEpE7gXNLVQDsAc/KVnnfkK4HeApwv\naSOwCTgtIionlT9KutJoJ9I5hOJ5BDMza4I+k0BETK5SfHkP894A3NDDtA7gVf2KzszMtir/YtjM\nrMScBMzMSsxJwMysxJwEzMxKzEnAzKzEnATMzErMScDMrMScBMzMSsxJwMysxJwEzMxKzEnAzKzE\nnATMzErMScDMrMScBMzMSsxJwMysxJwEzMxKzEnAzKzEakoCkmZKWiNpUaFsd0lzJN2f/w7P5ZJ0\nqaRlkhZIel1hmSl5/vvzQPVmZtZEtbYEZgFHdyubBsyNiPHA3Pwa4BjSAPPjganAdEhJgzQ+8RuA\ng4FzK4nDzMyao6YkEBG3A2u7FZ8IXJGfXwGcVCi/MpI7gN0kjQSOAuZExNqIeBSYw5aJxczMBlE9\n5wT2jIhVAPnvHrl8FPBQYb7OXNZT+RYkTZXUIamjq6urjhDNzKw3W+PEsKqURS/lWxZGzIiI9oho\nb2tra2hwZmb2vHqSwOrczUP+uyaXdwJjCvONBlb2Um5mZk1STxKYDVSu8JkC3FQoPyVfJXQI8Hju\nLvoZcKSk4fmE8JG5zMzMmmS7WmaSdA1wGDBCUifpKp8LgesknQo8CLw3z34LcCywDHgK+BBARKyV\n9AXgrjzf+RHR/WSzmZkNopqSQERM7mHS4VXmDeD0HuqZCcysOTozM9uq/IthM7MSq6klYNbKxk77\ncUPrW3HhcQ2tz6yVuSVgZlZiTgJmZiXmJGBmVmJOAmZmJeYkYGZWYk4CZmYl5iRgZlZiTgJmZiXm\nJGBmVmJOAmZmJeYkYGZWYk4CZmYl5iRgZlZiTgJmZiXmJGBmVmIDTgKS9pM0v/B4QtInJJ0n6eFC\n+bGFZc6WtEzSUklHNWYTzMxsoAY8qExELAUmAkgaBjwM3EgaU/iSiPj34vySJgCTgAOAvYBbJe0b\nEZsGGoOZmdWnUd1BhwMPRMSfepnnRODaiNgQEX8kDUR/cIPWb2ZmA9CoJDAJuKbw+gxJCyTNlDQ8\nl40CHirM05nLtiBpqqQOSR1dXV0NCtHMzLqrOwlI2h44AfhBLpoOvILUVbQKuKgya5XFo1qdETEj\nItojor2tra3eEM3MrAeNaAkcA9wTEasBImJ1RGyKiOeAb/F8l08nMKaw3GhgZQPWb2ZmA9SIJDCZ\nQleQpJGFae8CFuXns4FJknaQtA8wHrizAes3M7MBGvDVQQCSXgS8A/hIofirkiaSunpWVKZFxGJJ\n1wH3ARuB031lkJlZc9WVBCLiKeCl3co+0Mv8FwAX1LNOMzNrHP9i2MysxJwEzMxKzEnAzKzEnATM\nzErMScDMrMScBMzMSsxJwMysxJwEzMxKzEnAzKzEnATMzErMScDMrMScBMzMSsxJwMysxOq6i6jZ\nUDR22o8bWt+KC49raH1mjeSWgJlZiTkJmJmVmJOAmVmJ1Z0EJK2QtFDSfEkduWx3SXMk3Z//Ds/l\nknSppGWSFkh6Xb3rNzOzgWtUS+BtETExItrz62nA3IgYD8zNrwGOIQ0wPx6YCkxv0PrNzGwAtlZ3\n0InAFfn5FcBJhfIrI7kD2E3SyK0Ug5mZ9aERl4gG8HNJAfxnRMwA9oyIVQARsUrSHnneUcBDhWU7\nc9mqYoWSppJaCuy9994NCNGseXzJqbWyRiSBQyNiZf6inyPp973MqyplsUVBSiQzANrb27eYbmZm\njVF3d1BErMx/1wA3AgcDqyvdPPnvmjx7JzCmsPhoYGW9MZiZ2cDUlQQkvVjSLpXnwJHAImA2MCXP\nNgW4KT+fDZySrxI6BHi80m1kZmaDr97uoD2BGyVV6ro6In4q6S7gOkmnAg8C783z3wIcCywDngI+\nVOf6zcysDnUlgYhYDry2SvlfgMOrlAdwej3rNDOzxvEvhs3MSsxJwMysxJwEzMxKzEnAzKzEnATM\nzErMI4uZbWMafRsK8K0oyswtATOzEnMSMDMrMScBM7MScxIwMysxJwEzsxIb0lcHbY2rKMysbx5I\nZ9sxpJOAmdXGB0zl5e4gM7MScxIwMysxdweZWcvzr6S3HrcEzMxKbMBJQNIYSb+QtETSYkkfz+Xn\nSXpY0vz8OLawzNmSlklaKumoRmyAmZkNXD3dQRuBsyLinjzY/N2S5uRpl0TEvxdnljQBmAQcAOwF\n3Cpp34jYVEcMZmZWhwG3BCJiVUTck5+vA5YAo3pZ5ETg2ojYEBF/JA02f/BA129mZvVryDkBSWOB\nA4F5uegMSQskzZQ0PJeNAh4qLNZJD0lD0lRJHZI6urq6GhGimZlVUXcSkLQzcAPwiYh4ApgOvAKY\nCKwCLqrMWmXxqFZnRMyIiPaIaG9ra6s3RDMz60FdSUDSC0kJ4KqI+CFARKyOiE0R8RzwLZ7v8ukE\nxhQWHw2srGf9ZmZWn3quDhJwObAkIi4ulI8szPYuYFF+PhuYJGkHSfsA44E7B7p+MzOrXz1XBx0K\nfABYKGl+LvsMMFnSRFJXzwrgIwARsVjSdcB9pCuLTveVQWbWLL7JXTLgJBARv6Z6P/8tvSxzAXDB\nQNdpZmaN5dtGmJk1wLbasvBtI8zMSsxJwMysxJwEzMxKzEnAzKzEnATMzErMScDMrMScBMzMSsxJ\nwMysxJwEzMxKzEnAzKzEnATMzErMScDMrMScBMzMSsxJwMysxJwEzMxKzEnAzKzEBj0JSDpa0lJJ\nyyRNG+z1m5nZ8wY1CUgaBvwHcAwwgTQe8YTBjMHMzJ432C2Bg4FlEbE8Ip4BrgVOHOQYzMwsG+wx\nhkcBDxVedwJv6D6TpKnA1PxyvaSlA1zfCOCRAS47GFo9PnCMjdDq8UHrx9jq8UGDY9RX6lr85bXO\nONhJQFXKYouCiBnAjLpXJnVERHu99WwtrR4fOMZGaPX4oPVjbPX4YNuIsZrB7g7qBMYUXo8GVg5y\nDGZmlg12ErgLGC9pH0nbA5OA2YMcg5mZZYPaHRQRGyWdAfwMGAbMjIjFW3GVdXcpbWWtHh84xkZo\n9fig9WNs9fhg24hxC4rYokvezMxKwr8YNjMrMScBM7MSG5JJoBVvTSFppqQ1khYVynaXNEfS/fnv\n8CbGN0bSLyQtkbRY0sdbMMYdJd0p6d4c4+dz+T6S5uUYv58vOmgaScMk/U7SzS0a3wpJCyXNl9SR\ny1pmP+d4dpN0vaTf58/kG1slRkn75feu8nhC0idaJb7+GnJJoIVvTTELOLpb2TRgbkSMB+bm182y\nETgrIvYHDgFOz+9bK8W4AXh7RLwWmAgcLekQ4CvAJTnGR4FTmxgjwMeBJYXXrRYfwNsiYmLhuvZW\n2s8AXwd+GhF/B7yW9H62RIwRsTS/dxOBg4CngBtbJb5+i4gh9QDeCPys8Pps4Oxmx5VjGQssKrxe\nCozMz0cCS5sdYyG2m4B3tGqMwIuAe0i/OH8E2K7a/m9CXKNJXwBvB24m/UCyZeLLMawARnQra5n9\nDOwK/JF84UorxliI6Ujgf1o1vloeQ64lQPVbU4xqUix92TMiVgHkv3s0OR4AJI0FDgTm0WIx5q6W\n+cAaYA7wAPBYRGzMszR7f/8/4NPAc/n1S2mt+CD9Sv/nku7Ot2iB1trP44Au4Du5W+3bkl7cYjFW\nTAKuyc9bMb4+DcUkUNOtKaw6STsDNwCfiIgnmh1PdxGxKVIzfDTphoT7V5ttcKNKJB0PrImIu4vF\nVWZt9ufx0Ih4HanL9HRJb2lyPN1tB7wOmB4RBwJP0oJdK/nczgnAD5odSz2GYhLYlm5NsVrSSID8\nd00zg5H0QlICuCoifpiLWyrGioh4DLiNdP5iN0mVHz42c38fCpwgaQXpDrlvJ7UMWiU+ACJiZf67\nhtSXfTCttZ87gc6ImJdfX09KCq0UI6Qkek9ErM6vWy2+mgzFJLAt3ZpiNjAlP59C6odvCkkCLgeW\nRMTFhUmtFGObpN3y852AI0gnDH8BvCfP1rQYI+LsiBgdEWNJn7v/joiTWyU+AEkvlrRL5TmpT3sR\nLbSfI+LPwEOS9stFhwP30UIxZpN5visIWi++2jT7pMRWOllzLPAHUn/xOc2OJ8d0DbAKeJZ0pHMq\nqb94LnB//rt7E+N7E6mbYgEwPz+ObbEYXwP8Lse4CPhcLh8H3AksIzXNd2iB/X0YcHOrxZdjuTc/\nFlf+P1ppP+d4JgIdeV//FzC8lWIkXZjwF+AlhbKWia8/D982wsysxIZid5CZmdXIScDMrMScBMzM\nSsxJwMysxJwEzMxKzEnAtmmSNuU7OS7Odxc9U1LDPteSPihpr8LrbzfqhoSSTpL0uX4uc+u2cndK\n2zb4ElHbpklaHxE75+d7AFeTbuh1bj/qGBYRm3qYdhvwqYjoaES83er+DXBCRDzSj2WmAKMj4oJG\nx2Pl5JaADRmRboMwFThDyQclfaMyXdLNkg7Lz9dLOl/SPOCNkj4n6S5JiyTNyMu/B2gHrsqtjZ0k\n3SapPdcxOd+Xf5GkrxTWs17SBbllcoekPbvHKmlfYEMlAUiaJWm60pgOyyW9VWkMiiWSZhUWnU36\npapZQzgJ2JASEctJn+u+7uD4YtJtvd8QEb8GvhERr4+IVwE7AcdHxPWkX62eHOn+8U9XFs5dRF8h\n3R9oIvB6SScV6r4j0rgHtwP/p8r6DyXdCrtoeK7vk8CPgEuAA4BXS5qYt+9RYAdJL63h7TDrk5OA\nDUXV7tzZ3SbSzfIq3qY0+tdC0hfxAX0s/3rgtojoinSb6KuAyt04nyGNJQBwN2kcie5Gkm6XXPSj\nSP2zC4HVEbEwIp4j3d6hWMcaYC/MGmC7vmcx23ZIGkf6gl9DGi2teKCzY+H5XyvnASTtCHwTaI+I\nhySd123eqqvqZdqz8fzJtk1U/z97GnhJt7IN+e9zheeV18U6dszLm9XNLQEbMiS1AZeRunaCNILW\nREkvkDSGdMvkaipf+I/k8RTeU5i2DtilyjLzgLdKGpGHNJ0M/LIf4S4BXtmP+YG/3e31ZaRtM6ub\nWwK2rdspjzT2QtKR/3eByq2w/4c0TOFC0l1Hu/fBA2lsAknfyvOtIN2OvGIWcJmkp0lDQ1aWWSXp\nbNJtogXcEhH9uXXw7cBFklRoNdTiINL5ho19zmlWA18iatYkkr5OOg9waz+XmR0Rc7deZFYm7g4y\na54vke5L3x+LnACskdwSMDMrMbcEzMxKzEnAzKzEnATMzErMScDMrMScBMzMSux/AR6jX9gNJFv8\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112f6e668>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "\n",
    "## 使用本框及新框来回答问题 5##\n",
    "%matplotlib inline \n",
    "\n",
    "# 直方图示例，数据来自湾区样本\n",
    "sub_data = []\n",
    "cus_data = []\n",
    "\n",
    "with open('./data/Washington-2016-Summary.csv') as file:\n",
    "    reader = csv.DictReader(file)\n",
    "    for row in reader:\n",
    "        duration = float(row['duration'])\n",
    "\n",
    "        if row['user_type'] == 'Subscriber':\n",
    "            sub_data.append(float(row['duration']))\n",
    "        else:\n",
    "            cus_data.append(float(row['duration']))\n",
    "\n",
    "plt.hist(sub_data,bins = range(0,80,5))\n",
    "plt.title('Distribution of Trip Durations(Subscriber)')\n",
    "plt.xlabel('Duration (m)')\n",
    "\n",
    "plt.show()\n",
    "\n",
    "plt.hist(cus_data,bins = range(0,80,5))\n",
    "plt.title('Distribution of Trip Durations(Customer)')\n",
    "plt.xlabel('Duration (m)')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='eda_continued'></a>\n",
    "## 进行你自己的分析\n",
    "\n",
    "到目前为止，你已经对手头的数据进行了初步的探索，对比了三座美国城市的相应骑行量，以及会员和散客的骑行比例，你还选取了其中一座城市，从一般骑行时长的角度来研究会员和散客的区别，接下来的探索，你就要自己选择研究方向了，以下是一些探索方向的建议：\n",
    "\n",
    "- 不同月份或季度的骑客量有什么区别？哪个月份/季度的骑客量最高？会员骑行量与散客骑行量之比会受月份或季度的影响吗？\n",
    "- 周末和工作日的系统用户骑行模式有何区别？哪一天会员更可能使用骑行系统？散客呢？平均骑行时长会受一周内不同日期的影响吗？\n",
    "- 一天内的哪个时候系统使用最频繁？会员和散客的使用模式有区别吗？\n",
    "\n",
    "若你在问题 1 回答中提出的任何问题与上述参考类似，那现在就是研究这些的好时机。作为研究的一部分，你需要创建可视化界面，如果不想创建直方图，你可以参考 [Pyplot 文件](https://matplotlib.org/devdocs/api/pyplot_summary.html)。值得一提的是，如果你想为分类变量的不同值绘制统计图（如城市、用户类型），柱形图会大有帮助。[`.bar()`文件页面](https://matplotlib.org/devdocs/api/_as_gen/matplotlib.pyplot.bar.html#matplotlib.pyplot.bar)底部具备相关示例链接，可供你参考，助你创建自己需要的统计图。\n",
    "\n",
    "**问题 6**：探索另一个可用数据回答的问题，继续研究。在下方写下你想探索的问题，你的研究至少应包括两个变量，对比两组数据并使用一个可视化界面。\n",
    "\n",
    "**答案**: 会员和散客的骑行高峰分别是几点。数据显示会员在18点左右为骑行高峰，而散客的骑行高峰在12-13点左右"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZUAAAEWCAYAAACufwpNAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ngg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x112f82a20>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAY8AAAEWCAYAAACe8xtsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\norb/D1RGq+Pu7FR2AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10a01eef0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "## 使用本框及新框来继续探索数据集。 ##\n",
    "## 一旦你进行了自己的探索，请写下你的发现 ##\n",
    "## 请将发现写在上方的 Markdown 框中。                                     ##\n",
    "def ride_time_interval(filename,user_type):\n",
    "    data = []\n",
    "    with open(filename) as file:\n",
    "        reader = csv.DictReader(file)\n",
    "        for row in reader:\n",
    "            if row['user_type'] == user_type:\n",
    "                hour = int(row['hour'])\n",
    "                data.append(hour)\n",
    "    return data\n",
    "filename = './data/Washington-2016-Summary.csv'\n",
    "data = ride_time_interval(filename,'Subscriber')\n",
    "plt.hist(data,range=(0,24))\n",
    "plt.title('Washington Subscriber Ride Peak')\n",
    "plt.xlabel('Subscriber start time(h)')\n",
    "plt.ylabel('Number Of People')\n",
    "plt.show()\n",
    "\n",
    "data = ride_time_interval(filename,'Customer')\n",
    "plt.hist(data,range=(0,24))\n",
    "plt.title('Washington Customer Ride Peak')\n",
    "plt.xlabel('Customer start time(h)')\n",
    "plt.ylabel('Number Of People')\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<a id='conclusions'></a>\n",
    "## 结论\n",
    "\n",
    "恭喜你完成了本项目！这只是数据分析流程的一个样本，囊括了提出问题、整理数据、探索数据三大步骤。通常来讲，数据分析流程进行到这一步，你就可以进行统计检验或为数据拟合模型、得出预测，从而得出结论了。虽然还有许多潜在分析可以进行，但仅靠我们提供的这些数据是不可能完成的。例如，我们没有研究具体的地点数据。最常用的停车点在哪里？最常用的路线是什么？再如，天气也可能对每日骑客量造成很大的影响。下雨或下雪的时候，骑客量会受多大的影响？会员或散客谁更易受天气的影响？\n",
    "\n",
    "**问题 7**：撇开共享单车数据不谈，想出一个你想应用数据科学技术的课题或领域。对于该课题，你想了解些什么？\n",
    "\n",
    "**答案**：我想讲数据分析应用于健康领域。需要收集人群的各项健康指标（如体重、血压、体脂等），并根据健康状态分类，从中分析出各项指标对健康的影响。\n",
    "\n",
    "> **小贴士**：如果我们想同他人分享分析成果，我们也不一定要给他们一份 jupyter Notebook  (.ipynb) 文件的副本，我们可以将 Notebook 的输出结果以某种格式输出，这样一来，就算是没有安装 Python 的人也可以打开文件。在左上方的 **文件** 菜单中找到  **下载为**  子菜单，你可以在其中选择各种更加普遍的格式，比如 HTML (.html) 或 PDF (.pdf)。你可能还需要额外的包或软件来实现这类输出。\n",
    "\n",
    "> 如果你是通过教室的 Project Notebook 页面来进行本项目的，你也可以直接从工作空间提交本项目。 **但在此之前**，你得执行下方的代码框，为完成好的项目保存一份 HTML 的副本，并将其放到工作空间中。如果操作顺利，输出代码应为 0， 如果你点击左上角的 jupyter 图标，你就会看到工作空间的目录中出现了你的 .html 文件；或者你也可以根据上述段落提及的步骤来下载报告的 .html 格式副本，然后  _上传_ 该报告到目录之中（点击 jupyter 图标来实现这一步）。 \n",
    "\n",
    "> 无论是哪个办法，一旦你在工作空间中获取了 .html 格式的报告，你就可以点击工作空间右下方的“项目提交”按钮来完成提交。"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "255"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from subprocess import call\n",
    "call(['python', '-m', 'nbconvert', 'Bike_Share_Analysis.ipynb'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}