01_NumPy和陣列導向的程式設計-checkpoint.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "source": [
    "# NumPy 和陣列導向的程式設計\n",
    "\n",
    "`NumPy` 可以說是 Python 中最最標準的科學計算、數據分析套件。也因為 `NumPy` 的出現, 讓 Python 有了非常好的數據分析基礎, 一直到現在成為數據分析覇主。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "source": [
    "import numpy as np\r\n",
    "import pandas as pd"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 1. 陣列導向 101\n",
    "\n",
    "科學計算一個很核心的概念叫 \"array oriented\" 的寫法。Array 是 `numpy` 標準的資料結構, 和 list 很像, 但就差了那麼一點點。而這一點點讓我們在計算上是無比的方便。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【暖身】 計算平均\r\n",
    "\r\n",
    "某位同學期中考各科成績如下, 請幫他計算成績。\r\n",
    "\r\n",
    "    grades = [77, 85, 56, 90, 66]\r\n",
    "    \r\n",
    "請計算平均。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "source": [
    "# grades = [77, 85, 56, 90, 66]\r\n",
    "data = np.array([77, 85, 56, 90, 66])\r\n",
    "average = data.mean()\r\n",
    "sum = data.sum()\r\n",
    "print(\"average = \", average)\r\n",
    "print(\"sum = \", sum)\r\n"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "average =  74.8\n",
      "sum =  374\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【示範】陣列導向"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "source": [
    "print(average)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "74.8\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "最大值"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "source": [
    "data.max()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "90"
      ]
     },
     "metadata": {},
     "execution_count": 7
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "標準差"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "source": [
    "data.std()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "12.416118556135006"
      ]
     },
     "metadata": {},
     "execution_count": 10
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【暖身】 換算匯率\n",
    "\n",
    "假設今天我想查查號稱 Pentax 三公主的 31mm, 43mm, 77mm 三隻 limited 鏡頭在美國賣多少。於是我去 B&H 查了他們的價格分別是:\n",
    "\n",
    "    prices = [1096.95, 596.95, 896.95]\n",
    "    \n",
    "我又查了 Google 匯率 1 美金為 31.71 元。請把三支鏡頭的價格換算為台幣。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "source": [
    "prices = [1096.95, 596.95, 896.95]\r\n",
    "prices2=np.array(prices)\r\n",
    "print(prices2*31.71)\r\n",
    "\r\n",
    "\r\n"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "[34784.2845 18929.2845 28442.2845]\n"
     ]
    }
   ],
   "metadata": {
    "collapsed": true
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "先不管這實在有夠醜的數字, 我們要記得在科學計算中:\n",
    "\n",
    "### 儘可能不要使用迴圈\n",
    "\n",
    "這可能嗎?"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【示範】陣列換算匯率"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([34784.2845, 18929.2845, 28442.2845])"
      ]
     },
     "metadata": {},
     "execution_count": 15
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "哦哦, 傑克, 這太神奇了!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 2. 其實 array 還有很多功能\n",
    "\n",
    "### 【練習】 成績計算\n",
    "\n",
    "一位老師成績這樣算的:\n",
    "\n",
    "* 平時成績 20%\n",
    "* 期中考   35%\n",
    "* 期未考   45%\n",
    "\n",
    "有位同學\n",
    "\n",
    "* 平時成績 85 分\n",
    "* 期中 70 分\n",
    "* 期末 80 分\n",
    "\n",
    "這位同學的學期成績是多少?"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "source": [
    "grades = np.array([85,70,80])\r\n",
    "weights = np.array([0.2, 0.35, 0.45])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "這還不是我們要的最終成績啊!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "source": [
    "weighted_grades=grades*weights\r\n",
    "print(weighted_grades)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "[17.  24.5 36. ]\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【提示】 array 還有很多函數可以用\r\n",
    "\r\n",
    "可以先打入\r\n",
    "\r\n",
    "    weighted_grades.\r\n",
    "    \r\n",
    "先不要按 `enter` 或 `shift-enter`, 而是按 `tab`...\r\n"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "source": [
    "weighted_grades.sum()  "
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "77.5"
      ]
     },
     "metadata": {},
     "execution_count": 21
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 一行完成成績計算"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "source": [
    "weighted_grades.sum()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "77.5"
      ]
     },
     "metadata": {},
     "execution_count": 22
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 3. 重要的 array 大變身!\n",
    "\n",
    "我們在數據分析, 常常要改 array 的型式。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【練習】 一個 50 個數字的 array\n",
    "\n",
    "先想辦法、用亂數做出 50 個數字的 array, 叫做 A 好了。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "source": [
    "from numpy import random \r\n",
    "\r\n",
    "A = random.randint(1, 50, 50)\r\n",
    "\r\n",
    "print(A)\r\n"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "[28 46  2 20 33 47  5 34 11 20 48 33 18 18 37  3 43 46  1 34 47  4  1 12\n",
      " 43 21 41  1 43 13 28 38 31 33 27  8 38  2 44 39 36 35 48 42 17 49 29 48\n",
      " 21 33]\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "0.737223397267063"
      ]
     },
     "metadata": {},
     "execution_count": 30
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 檢查 A 的 `shape`"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "source": [
    "B.shape"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "(50,)"
      ]
     },
     "metadata": {},
     "execution_count": 81
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 更改 A 的 shape"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "source": [
    "B.reshape((5,5,2))"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([[[26, 40],\n",
       "        [17, 19],\n",
       "        [15, 45],\n",
       "        [42, 29],\n",
       "        [38,  3]],\n",
       "\n",
       "       [[ 9, 13],\n",
       "        [ 5, 38],\n",
       "        [43, 17],\n",
       "        [25, 19],\n",
       "        [24, 49]],\n",
       "\n",
       "       [[16, 48],\n",
       "        [22, 49],\n",
       "        [ 5, 29],\n",
       "        [43, 25],\n",
       "        [18, 38]],\n",
       "\n",
       "       [[13, 30],\n",
       "        [21, 15],\n",
       "        [21, 21],\n",
       "        [45, 28],\n",
       "        [27,  2]],\n",
       "\n",
       "       [[26, 35],\n",
       "        [38, 28],\n",
       "        [ 1, 24],\n",
       "        [46,  2],\n",
       "        [39, 27]]])"
      ]
     },
     "metadata": {},
     "execution_count": 84
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 也可以用 `reshape`\n",
    "\n",
    "但要注意, reshape 並沒有改原來的陣列。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([[0.7372234 , 0.6126442 , 0.27626397, 0.28984569, 0.4988311 ],\n",
       "       [0.42021495, 0.4515489 , 0.15328583, 0.93455268, 0.23537373],\n",
       "       [0.24761985, 0.41945762, 0.44871061, 0.66726833, 0.15622678],\n",
       "       [0.70315927, 0.98138833, 0.8233088 , 0.85394749, 0.14797509],\n",
       "       [0.32816167, 0.64607213, 0.96849395, 0.90799427, 0.19579351],\n",
       "       [0.35618479, 0.30311169, 0.32568614, 0.66885309, 0.67844806],\n",
       "       [0.18201376, 0.69592441, 0.11077285, 0.22884185, 0.29488536],\n",
       "       [0.90457337, 0.19872841, 0.38361489, 0.35731267, 0.76200102],\n",
       "       [0.92178147, 0.44190395, 0.07941469, 0.03253205, 0.44805018],\n",
       "       [0.38885104, 0.15473767, 0.55240742, 0.80182196, 0.70921062]])"
      ]
     },
     "metadata": {},
     "execution_count": 35
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 拉平 `ravel`\n",
    "\n",
    "雖然你想一想就知道可以用 `shape` 或 `reshape` 把多維陣列拉成一維。不過用 `ravel` 很潮。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "source": [
    "A.ravel()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([28, 46,  2, 20, 33, 47,  5, 34, 11, 20, 48, 33, 18, 18, 37,  3, 43,\n",
       "       46,  1, 34, 47,  4,  1, 12, 43, 21, 41,  1, 43, 13, 28, 38, 31, 33,\n",
       "       27,  8, 38,  2, 44, 39, 36, 35, 48, 42, 17, 49, 29, 48, 21, 33])"
      ]
     },
     "metadata": {},
     "execution_count": 86
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 4. 快速 array 生成法"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 都是 0 的 array"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "source": [
    "np.zeros(50)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.,\n",
       "       0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.])"
      ]
     },
     "metadata": {},
     "execution_count": 90
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】 都是 1 的 array"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 91,
   "source": [
    "np.ones(50)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,\n",
       "       1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.,\n",
       "       1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1., 1.])"
      ]
     },
     "metadata": {},
     "execution_count": 91
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】單位矩陣"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 96,
   "source": [
    "np.eye(5) "
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([[1., 0., 0., 0., 0.],\n",
       "       [0., 1., 0., 0., 0.],\n",
       "       [0., 0., 1., 0., 0.],\n",
       "       [0., 0., 0., 1., 0.],\n",
       "       [0., 0., 0., 0., 1.]])"
      ]
     },
     "metadata": {},
     "execution_count": 96
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】給定範圍均勻生出 n 個點"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 98,
   "source": [
    "random.randint(1,50,50)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([14,  5,  9, 30,  3, 41, 29, 38, 12,  8, 35, 26, 34,  9, 21, 41, 44,\n",
       "       25, 37, 30,  2, 33, 46,  7, 11, 11, 49, 23,  1,  6, 23, 37,  5, 14,\n",
       "        1, 38, 13, 37, 14,  1, 35, 39, 14, 15, 10, 13, 37, 31, 43,  3])"
      ]
     },
     "metadata": {},
     "execution_count": 98
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】`range` 的 array 版\n",
    "\n",
    "就是 `arange`。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "source": [
    "np.arange(1,10)"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([1, 2, 3, 4, 5, 6, 7, 8, 9])"
      ]
     },
     "metadata": {},
     "execution_count": 101
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 5. 超重要 `axis` 觀念\n",
    "\n",
    "初學 `numpy` 很多人有點弄不清楚 `axis` 概念。其實掌握矩陣, 或很像矩陣的陣列都是「先列後行」就可以!\n",
    "\n",
    "我們先弄個 array 來練習。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 141,
   "source": [
    "np.array([1,2,3,4,5,6])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([1, 2, 3, 4, 5, 6])"
      ]
     },
     "metadata": {},
     "execution_count": 141
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【重點】 一列一列算下來是 `axis=0`"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "![axis=0](images/axis0.png)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 137,
   "source": [
    "import numpy as np\r\n",
    "np.array([5,7,9,11,13])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([ 5,  7,  9, 11, 13])"
      ]
     },
     "metadata": {},
     "execution_count": 137
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【重點】 一行一行算過去是 `axis=1`\n",
    "\n",
    "![axis=1](images/axis1.png)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 142,
   "source": [
    "np.array([5,7,9,11,13])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([ 5,  7,  9, 11, 13])"
      ]
     },
     "metadata": {},
     "execution_count": 142
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【提示】當然也有可能全部算"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "source": [
    "A.sum()"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "1399"
      ]
     },
     "metadata": {},
     "execution_count": 143
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 6. array 過濾器\n",
    "\n",
    "篩出我們要的資料, 這樣的技巧非常重要!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【例子】篩出大於 0 的數\n",
    "\n",
    "我們有個陣列, 想找出大於 0 的數。<br>\n",
    "L = np.array([3, -2, -1, 5, 7, -3])"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "source": [
    "L = np.array([3, -2, -1, 5, 7, -3])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "我們可以很白痴的自己判斷...<br>\n",
    "c = np.array([True,False,False,True,True,False])"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "source": [
    "c = np.array([True,False,False,True,True,False])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "這是做啥呢? 我們可以瞬間..."
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "source": [
    "print(L>0)"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "[ True False False  True  True False]\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "除了自己做很白痴, 這看來很厲害!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "事實上我們可以叫 `numpy` 做!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([ True, False, False,  True,  True, False])"
      ]
     },
     "metadata": {},
     "execution_count": 61
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "這有點強, 我們還可以一次到位!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([3, 5, 7])"
      ]
     },
     "metadata": {},
     "execution_count": 64
    }
   ],
   "metadata": {
    "scrolled": true
   }
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 7. 次元切割刀\n",
    "\n",
    "`numpy` 中 array 的切割法和 list 很像。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "source": [
    "np.array([0,1,2,3,4,5,6,7,8,9])"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([0, 1, 2, 3, 4, 5, 6, 7, 8, 9])"
      ]
     },
     "metadata": {},
     "execution_count": 147
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "source": [
    "np.array([0,1,2,3,4,5,6,7,8,9])[4:7]"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([4, 5, 6])"
      ]
     },
     "metadata": {},
     "execution_count": 148
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【技巧】2維陣列切法\n",
    "\n",
    "記得先列後行!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "source": [
    "A[1:2]"
   ],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([46])"
      ]
     },
     "metadata": {},
     "execution_count": 162
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "要所有的row, 切出行 1-3 位置。"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([[1, 2],\n",
       "       [6, 7]])"
      ]
     },
     "metadata": {},
     "execution_count": 70
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "要所有的行col, 切出第 1 列row!"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "source": [],
   "outputs": [
    {
     "output_type": "execute_result",
     "data": {
      "text/plain": [
       "array([5, 6, 7, 8, 9])"
      ]
     },
     "metadata": {},
     "execution_count": 71
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "## 8. `NumPy` 的 `zip` 和 `unzip`"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "之前我們介紹 list 可以用 `zip` 和 `unzip` (其實還是 `zip`) 做到的資料格式變換, 在 array 中怎麼做呢?\n",
    "\n",
    "![zip and unzip](images/zip.png)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【重點】array 的 `zip`\n",
    "\n",
    "![array zip](images/arrzip.png)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "source": [
    "x = np.array([1,2,3,4])\r\n",
    "y = np.array([5,6,7,8])"
   ],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "source": [
    " z = zip(x,y)\r\n",
    " print(list(z))"
   ],
   "outputs": [
    {
     "output_type": "stream",
     "name": "stdout",
     "text": [
      "[(1, 5), (2, 6), (3, 7), (4, 8)]\n"
     ]
    }
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [],
   "outputs": [],
   "metadata": {}
  },
  {
   "cell_type": "markdown",
   "source": [
    "### 【重點】array 的 `unzip`\n",
    "\n",
    "這裡其實只需要用到 array 的切割法...\n",
    "\n",
    "![array zip](images/arrunzip.png)"
   ],
   "metadata": {}
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "source": [],
   "outputs": [],
   "metadata": {
    "collapsed": true
   }
  }
 ],
 "metadata": {
  "kernelspec": {
   "name": "python3",
   "display_name": "Python 3.7.6 64-bit ('base': conda)"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  },
  "interpreter": {
   "hash": "513dc2e41d739bb2c947903f3c0bbf636d03aa53ab50e61c694a27481c81805e"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}