Lab_python_matplotlib_numpy_pandas.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Bài tập bổ trợ\n",
    "## MaSSP 2018, Computer Science"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Trong file này, các bạn sẽ vận dụng những hàm đã học trong những notebook trước vào những dữ liệu thực tế. Đáp án của những bài tập này có thể được tìm thấy ở cuối notebook, nhưng các bạn chỉ nên tham khảo sau khi đã giải và kiểm tra cẩn thận kết quả của mình. Mọi thắc mắc hay phát hiện lỗi sai, hãy post lên Facebook group: [MaSSP Tin](https://www.facebook.com/groups/183191748970952/)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import numpy as np\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Python\n",
    "Sau đây là một số bài tập giúp các bạn làm quen hơn với ngôn ngữ lập trình Python về mặt cú pháp và cơ sở dữ liệu. Nếu các bạn đã sử dụng ngôn ngữ này thành thạo có thể bỏ qua phần này."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.1 Sử dụng list\n",
    "Cho một `list` sau:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "percents = [20.7, 19.4, 14.7, 7.6, 6.9]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Như các bạn đã biết, các hàm `min, max` có thể được áp dụng lên `list` để tìm ra giá trị nhỏ nhất và lớn nhất của `list`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "6.9\n"
     ]
    }
   ],
   "source": [
    "print(min(percents))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1.1.1. Bằng cách sử dụng vòng lặp `for`, hãy viết các hàm `find_min, find_max` để tìm ra giá trị nhỏ nhất và lớn nhất của một `list`. Sau khi hoàn thiện, các bạn hãy dùng các hàm `min, max` để kiểm tra kết quả của mình."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def find_min(array):\n",
    "    pass\n",
    "\n",
    "def find_max(array):\n",
    "    pass\n",
    "\n",
    "# kiểm tra kết quả\n",
    "assert min(percents) == find_min(percents)\n",
    "assert max(percents) == find_min(percents)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1.1.2. Tương tự, hãy dùng vòng lặp `for` để hoàn thiện hàm `find_mean` dùng để tính giá trị trung bình của một `list` các số."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def find_mean(array):\n",
    "    pass\n",
    "\n",
    "# kiểm tra kết quả\n",
    "asset find_mean(percents) == 13.86"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Mảng `percents` chúng ta định nghĩa ở trên thực ra là phần trăm thị phần của các công ty sản xuất máy tính lớn trên thế giới vào năm 2016 (nguồn [Wikipedia](https://en.wikipedia.org/wiki/Market_share_of_personal_computer_vendors#Top_vendors_market_share_(2017)). Tên của các công ty được lưu lại trong mảng `companies`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "percents = [20.7, 19.4, 14.7, 7.6, 6.9]\n",
    "companies = ['Lenovo', 'HP', 'Dell', 'Asus', 'Apple']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1.1.3. Như các bạn để ý trong bảng số liệu trên trang Wikipedia, 5 hãng máy tính này có thị phần cao nhất, và những công ty khác chỉ chiếm 30.7% thị phần. Hãy thêm vào `percents` và `companies` lần lượt các số `30.7` và xâu 'Others'."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1.2 Sử dụng dict\n",
    "Vị trí của tên mỗi công ty trong `companies` tương ứng với vị trí của thị phần của công ty đó trong `percents`, ví dụ 'Apple' chiếm 6.9% thị phần.\n",
    "\n",
    "1.2.1. Hãy khởi tạo một `dict` có tên `company_to_percent`, trong đó mỗi key là tên của một công ty hoặc 'Others', và value của mỗi key ứng với thị phần của công ty đó."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n",
    "company_to_percent = ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1.2.2. Sử dụng biến `company_to_percent` ta vừa định nghĩa ở trên, hãy cho biết thị phần của công ty 'Dell'?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1.2.3. Giả sử có sai sót trong việc thu thập dữ liệu, công ty 'Lenovo' chiếm đến 21.5% thị phần thay vì 20.7%, và công ty 'Dell' chiếm 14.5% thay vì 14.7%. Bạn hãy cập nhật lại những thông tin này trong `company_to_percent`, và nhớ là thị phần của tất cả các công ty phải bằng 100%!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Numpy và matplotlib\n",
    "Qua bài tập này, chúng ta sẽ thấy rõ lợi ích của việc sử dụng các thư viện có sẵn khi làm việc với nhiều dữ liệu. Thay vì phải viết vòng lặp, có rất nhiều hàm có sẵn trong những thư viện này giúp ta thực hiện nhanh chóng những thao tác thường gặp và dành thời gian giải quyết những vấn đề phức tạp hơn.\n",
    "\n",
    "### 2.1 Numpy\n",
    "Sau đây là 2 `list` chứa thị phần của 5 công ty trên trong năm 2016 và 2017:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "percents_2016 = [20.7, 19.4, 14.7, 7.6, 6.9]\n",
    "percents_2017 = [20.8, 21.0, 15.2, 6.8, 7.4]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2.1.1. Hãy khởi tạo 2 vectors `percents_2016_array, percents_2017_array` từ 2 `list` trên."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "percents_2016_array = ...\n",
    "percents_2017_array = ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2.1.2. Công ty nào bị giảm thị phần vào năm 2017? Gợi ý: dùng hàm [numpy.where](https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.where.html)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2.1.3. Gộp 2 vectors trên thành một array duy nhất, với hàng đầu tiên chứa thị phần năm 2016, và hàng thứ 2 chứa thị phần năm 2017."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "percents_array = ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2.1.4. Sử dụng hàm `numpy.mean()`, hãy trả lại các array chứa a) giá trị thị phần trung bình của cả 5 công ty trong từng năm 2016 và 2017, và b) giá trị thị phần trung bình của mỗi công ty trong 2 năm."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2 Matplotlib\n",
    "Bạn hãy vẽ một đồ thị tuỳ ý dựa trên số liệu trên Wikipedia, ví dụ thị phần của công ty Apple và Lenovo qua từng năm, và hãy nhớ đồ thị cần có tên và chú thích!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Pandas\n",
    "Khi làm bài tập, các bạn nên tham khảo các hàm thông dụng trong `cheatsheet.ipynb` và tra cứu thư viện pandas hoặc Stackoverflow khi có bug.\n",
    "\n",
    "Bộ dữ liệu sử dụng trong phần này được tải về từ trang https://dasl.datadescription.com/story/climate-change-2016/ và được chuyển hoá sang dạng `csv` (xem `climate-change-2016.csv`). Bạn hãy mở xem nội dung của tệp tin `climate-change-2016.csv` để hiểu ý nghĩa của từng cột dữ liệu.\n",
    "\n",
    "Trích dẫn từ trang web:\n",
    "> Scientists claim that changes in the mean global temperature are primarily due to changes in CO2 levels. Both trends are here from 1959 to 2016. For an alternative, the data includes the annual closing price of the Dow Jones Industrial Average. Can it predict global temperature?\n",
    "\n",
    "Tạm dịch là:\n",
    "> Các nhà khoa học cho rằng những thay đổi trong nhiệt độ trung bình toàn cầu có nguyên nhân chủ yếu do thay đổi của nồng độ CO2. Xu hướng biến đổi của cả hai thông số này từ năm 1959 đến năm 2016 được ghi lại tại đây. Dữ liệu bao gồm giá cuối năm của [Chỉ số trung bình công nghiệp Dow Jones](https://vi.wikipedia.org/wiki/Ch%E1%BB%89_s%E1%BB%91_trung_b%C3%ACnh_c%C3%B4ng_nghi%E1%BB%87p_Dow_Jones) như một cách giải thích thay thế khác. Liệu thông tin này có thể dự đoán được nhiệt độ trung bình toàn cầu?\n",
    "\n",
    "Câu hỏi ngoài lề: không nhìn vào số liệu, theo em chỉ số trung bình công nghiệp Dow Jones sẽ có mối liên hệ thế nào tới nồng độ CO2?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.1 Đọc dữ liệu\n",
    "3.1.1. Các bạn hãy load dữ liệu trong tệp tin `climate-change-2016.csv` vào biến `df`. Hãy nhớ tra cứu cheatsheet hoặc các bài giảng trước trước khi tham khảo gợi ý ở cuối notebook này!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n",
    "df = ..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3.1.2. Hãy in ra 10 hàng đầu tiên và 10 hàng cuối cùng của bảng dữ liệu này. Nhìn qua các con số, bạn có nhận xét gì về xu hướng của các thông số này? (tăng hay giảm, biên độ tăng giảm lớn hay nhỏ...)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "3.1.3. Trong bảng dữ liệu này có năm nào không có hoặc thiếu dữ liệu không?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.2 Minh hoạ dữ liệu trong Pandas\n",
    "3.2.1 Trong bài giảng về pandas, chúng ta đã học cách vẽ đồ thị từ DataFrame, trong đó mỗi cột dữ liệu được biểu diễn bằng một đường. Hãy sử dụng phương pháp đó để vẽ đồ thị biến thiên của các cột theo thời gian."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# code"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Bạn có nhận xét gì khi vẽ các chỉ số này trên cùng một đồ thị? Chú ý đến độ lớn và đơn vị của từng chỉ số.\n",
    "\n",
    "Có thể thấy rằng biên độ của nhiệt độ là không lớn, chỉ khoảng 1 độ, trong khi biên độ của chỉ số CO2 theo phần triệu là khoảng 100, và biên độ của chỉ số DJIA còn tới hàng chục nghìn. Khi thể hiện những chỉ số này trên cùng một đồ thị, ta rất dễ ngộ nhận rằng chúng không có mối liên hệ nào vì biên độ nhiệt độ trông giống như được giữ nguyên. Các bạn hãy đón xem các bài giảng tiếp theo để học các phương pháp xác định mối liên hệ giữa các thông số này nhé!\n",
    "\n",
    "Một lần nữa, tất cả thắc mắc, phát hiện lỗi sai, xin chia sẻ lên [Facebook group nhóm Tin học](https://www.facebook.com/groups/183191748970952/)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 3.3 Bài tập nâng cao\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Nếu có thời gian và không muốn đợi đến chương trình MaSSP, các bạn hãy tìm hiểu thêm các hàm hữu ích khác và cũng phức tạp hơn trong pandas như `groupby`, `size`, `agg` để thực hiện được nhiều thao tác hơn, ví dụ để trả lời một số câu hỏi liên quan đến thống kê trong [bộ quiz về các thế vận hội Olympics](https://www.theguardian.com/sport/2016/aug/05/rio-2016-big-olympics-quiz-games). Dữ liệu cần thiết đã có sẵn trong `medalists_1896_to_2008.csv`, chứa thông tin các thành phố đăng cai, năm diễn ra Olympics, tên các vận động viên, các môn thể thao, và loại huy chương từ năm 1896 đến 2008. Bạn có thể dùng các công cụ đã học để trả lời bộ quiz này không?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gợi ý\n",
    "Đây chỉ là gợi ý cách làm, có thể có những cách ngắn gọn hơn. Nếu bạn tìm thấy lỗi sai hoặc có cách làm ngắn gọn hơn, hãy chia sẻ lên Facebook group!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 1. Python"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 1.1.1\n",
    "def find_min(array):\n",
    "    min_value = None\n",
    "    for elt in array:\n",
    "        if not min_value or elt < min_value:\n",
    "            min_value = elt\n",
    "    return min_value\n",
    "\n",
    "def find_max(array):\n",
    "    max_value = None\n",
    "    for elt in array:\n",
    "        if not max_value or elt > max_value:\n",
    "            max_value = elt\n",
    "    return max_value"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 1.1.2\n",
    "# sử dụng hàm sum\n",
    "def find_mean(array):\n",
    "    if len(array) == 0:\n",
    "        return None\n",
    "    return sum(array) / len(array)\n",
    "\n",
    "# không sử dụng hàm sum\n",
    "def find_mean(array):\n",
    "    if len(array) == 0:\n",
    "        return None\n",
    "    sum_value = 0\n",
    "    for elt in array:\n",
    "        sum_value += elt\n",
    "    return elt / len(array)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 1.1.3\n",
    "percents = [20.7, 19.4, 14.7, 7.6, 6.9]\n",
    "companies = ['Lenovo', 'HP', 'Dell', 'Asus', 'Apple']\n",
    "percents.append(30.7)\n",
    "companies.append('Others')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Apple': 6.9,\n",
       " 'Asus': 7.6,\n",
       " 'Dell': 14.7,\n",
       " 'HP': 19.4,\n",
       " 'Lenovo': 20.7,\n",
       " 'Others': 30.7}"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1.2.1\n",
    "company_to_percent = { companies[i]:percents[i] for i in range(len(percents)) }\n",
    "company_to_percent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "14.7"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1.2.2\n",
    "company_to_percent['Dell']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'Apple': 6.9,\n",
       " 'Asus': 7.6,\n",
       " 'Dell': 14.5,\n",
       " 'HP': 19.4,\n",
       " 'Lenovo': 21.5,\n",
       " 'Others': 30.099999999999994}"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 1.2.3\n",
    "company_to_percent['Dell'] = 14.5\n",
    "company_to_percent['Lenovo'] = 21.5\n",
    "sum_without_others = 0\n",
    "for company in company_to_percent:\n",
    "    if company != 'Others':\n",
    "        sum_without_others += company_to_percent[company]\n",
    "company_to_percent['Others'] = 100 - sum_without_others\n",
    "company_to_percent"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Numpy và matplotlib"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 2.1.1\n",
    "percents_2016_array = np.array(percents_2016)\n",
    "percents_2017_array = np.array(percents_2017)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 0.1  1.6  0.5 -0.8  0.5]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "3"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 2.1.2\n",
    "diff_percents_array = percents_2017_array - percents_2016_array\n",
    "print(diff_percents_array)\n",
    "np.where(diff_percents_array < 0)[0][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[ 20.7,  19.4,  14.7,   7.6,   6.9],\n",
       "       [ 20.8,  21. ,  15.2,   6.8,   7.4]])"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 2.1.3\n",
    "percents_array = np.array([percents_2016_array, percents_2017_array])\n",
    "percents_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[ 20.75  20.2   14.95   7.2    7.15]\n",
      "[ 13.86  14.24]\n"
     ]
    }
   ],
   "source": [
    "# 2.1.4\n",
    "print(np.mean(percents_array, axis=0))\n",
    "print(np.mean(percents_array, axis=1))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZoAAAEJCAYAAACt9OGOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nhfr/TERF3By/X1QAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1108ad710>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 2.2\n",
    "# Phỏng theo https://matplotlib.org/gallery/api/barchart.html \n",
    "companies = ['Lenovo', 'HP', 'Dell', 'Asus', 'Apple']\n",
    "percents_array = np.array([percents_2016_array, percents_2017_array])\n",
    "\n",
    "ind = np.arange(len(companies))  # the x locations for the groups\n",
    "width = 0.35  # the width of the bars\n",
    "\n",
    "fig, ax = plt.subplots()\n",
    "rects1 = ax.bar(ind - width/2, percents_array[0], width,\n",
    "                color='SkyBlue', label='2016')\n",
    "rects2 = ax.bar(ind + width/2, percents_array[1], width,\n",
    "                color='IndianRed', label='2017')\n",
    "\n",
    "# Add some text for labels, title and custom x-axis tick labels, etc.\n",
    "ax.set_ylabel('Thị phần (%)')\n",
    "ax.set_title('Thị phần máy tính cá nhân của Lenovo, HP, Dell, Asus, Apple')\n",
    "ax.set_xticks(ind)\n",
    "ax.set_xticklabels(companies)\n",
    "ax.legend()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 3. Pandas\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# 3.1.1\n",
    "df = pd.read_csv('climate-change-2016.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "# 3.1.2\n",
    "df.head(10)\n",
    "df.tail(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 58 entries, 0 to 57\n",
      "Data columns (total 4 columns):\n",
      "year                   58 non-null int64\n",
      "Global_temp_anomaly    58 non-null float64\n",
      "CO2(ppm)               58 non-null float64\n",
      "DJIA                   58 non-null float64\n",
      "dtypes: float64(3), int64(1)\n",
      "memory usage: 1.9 KB\n"
     ]
    }
   ],
   "source": [
    "# 3.1.3\n",
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYAAAAEKCAYAAAAb7IIBAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nBAAhhJigJAAIIcQE9f8B4mbBBsfp+OoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10f72cf28>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 3.2.1\n",
    "df.plot(x='year', y='Global_temp_anomaly')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 92,
   "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>year</th>\n",
       "      <th>Global_temp_anomaly</th>\n",
       "      <th>CO2(ppm)</th>\n",
       "      <th>DJIA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1959</td>\n",
       "      <td>0.0596</td>\n",
       "      <td>315.97</td>\n",
       "      <td>679.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>1964</td>\n",
       "      <td>-0.1495</td>\n",
       "      <td>319.62</td>\n",
       "      <td>874.13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>1969</td>\n",
       "      <td>0.0929</td>\n",
       "      <td>324.62</td>\n",
       "      <td>800.36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>1974</td>\n",
       "      <td>-0.0719</td>\n",
       "      <td>330.18</td>\n",
       "      <td>616.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>1979</td>\n",
       "      <td>0.2273</td>\n",
       "      <td>336.84</td>\n",
       "      <td>838.74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>1984</td>\n",
       "      <td>0.1490</td>\n",
       "      <td>344.65</td>\n",
       "      <td>1211.57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>1989</td>\n",
       "      <td>0.2970</td>\n",
       "      <td>353.12</td>\n",
       "      <td>2753.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>1994</td>\n",
       "      <td>0.3409</td>\n",
       "      <td>358.83</td>\n",
       "      <td>3833.40</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>1999</td>\n",
       "      <td>0.4438</td>\n",
       "      <td>368.38</td>\n",
       "      <td>11497.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>2004</td>\n",
       "      <td>0.5783</td>\n",
       "      <td>377.52</td>\n",
       "      <td>10783.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>2009</td>\n",
       "      <td>0.6367</td>\n",
       "      <td>387.43</td>\n",
       "      <td>10428.10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>2014</td>\n",
       "      <td>0.7408</td>\n",
       "      <td>398.65</td>\n",
       "      <td>17823.07</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    year  Global_temp_anomaly  CO2(ppm)      DJIA\n",
       "0   1959               0.0596    315.97    679.36\n",
       "5   1964              -0.1495    319.62    874.13\n",
       "10  1969               0.0929    324.62    800.36\n",
       "15  1974              -0.0719    330.18    616.24\n",
       "20  1979               0.2273    336.84    838.74\n",
       "25  1984               0.1490    344.65   1211.57\n",
       "30  1989               0.2970    353.12   2753.20\n",
       "35  1994               0.3409    358.83   3833.40\n",
       "40  1999               0.4438    368.38  11497.10\n",
       "45  2004               0.5783    377.52  10783.00\n",
       "50  2009               0.6367    387.43  10428.10\n",
       "55  2014               0.7408    398.65  17823.07"
      ]
     },
     "execution_count": 92,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df5 = df[::5]\n",
    "df5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXQAAAD8CAYAAABn919SAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nl6RGGOiS1AgDXZIa8f/VkuxpwaMCowAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x110c91f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df5.Global_temp_anomaly.plot(kind='barh')\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.6.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}