wk11/MY470_wk11_class.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "### MY470 Computer Programming\n",
    "# Useful Python Modules and Libraries\n",
    "### Week 11 Lab"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Working with Dates: `datetime`\n",
    "\n",
    "https://docs.python.org/3/library/datetime.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'datetime.datetime'> 2002-02-25 15:51:15\n",
      "<class 'datetime.timedelta'> 1 day, 0:00:00\n",
      "<class 'datetime.timedelta'> 1 day, 0:00:00\n"
     ]
    }
   ],
   "source": [
    "from datetime import datetime, timedelta\n",
    "\n",
    "# Transform string into datetime object\n",
    "string = '2002-02-25 15:51:15'\n",
    "dt = datetime.strptime(string, \"%Y-%m-%d %H:%M:%S\")\n",
    "print(type(dt), dt)\n",
    "\n",
    "# Get difference between two dates\n",
    "day1 = dt - datetime(2002, 2, 24, 15, 51, 15)\n",
    "print(type(day1), day1)\n",
    "\n",
    "day2 = timedelta(hours=24)\n",
    "print(type(day2), day2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Saving Python Objects to Disk: `pickle`\n",
    "\n",
    "https://docs.python.org/3/library/pickle.html"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'dict'> {1: 'a', 2: 'b'}\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "\n",
    "# Note: Open files in binary mode\n",
    "\n",
    "# Pickle Python object\n",
    "obj = {1:'a', 2:'b'}\n",
    "with open('dict_pickle', 'wb') as fw:\n",
    "    pickle.dump(obj, fw)\n",
    "\n",
    "# Unpickle saved Python object\n",
    "with open('dict_pickle', 'rb') as fr:\n",
    "    obj_retrieved = pickle.load(fr)\n",
    "    \n",
    "print(type(obj_retrieved), obj_retrieved)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "outputs": [],
   "source": [
    "# Exercise 1: Read the file '../../data/enwiki_2002_sample.txt' \n",
    "# and save the data in a variable as a list. Make sure all data \n",
    "# are in the correct # data type, e.g., save dates as datetime objects.\n",
    "# Print the first 3 elements of the list to confirm.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "collapsed": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "# Exercise 2: Pickle the list variable from exercise 1. \n",
    "# Then load it back in memory from the pickle.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "# Useful Libraries for Computational Social Science and Social Data Science\n",
    "\n",
    "* NumPy\n",
    "* Pandas\n",
    "* StatsModels\n",
    "* NetworkX\n",
    "* Scikit-learn\n",
    "* Matplotlib (introduced in week 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Working with Arrays and Matrices: `numpy`\n",
    "\n",
    "![NumPy](figs/numpy.png \"NumPy\")\n",
    "\n",
    "* Python library for operating with large multi-dimensional arrays and matrices\n",
    "* Partially written in C, so optimized for performance\n",
    "* Offers fast numerical computations and high-level math functions\n",
    "* Provides functionality similar to MATLAB (especially together with `scipy` and `matplotlib`)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## NumPy Arrays\n",
    "\n",
    "* Multi-dimensional: 0 (scalar), 1 (vector), 2 (matrix), 3 (e.g., image), ...\n",
    "* Dense: all elements exist\n",
    "* Homogeneous: all elements are of the same type"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[1 2 3]\n",
      " [4 5 6]]\n",
      "<class 'numpy.ndarray'>\n",
      "2 (2, 3) int64\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "\n",
    "a = np.array([[1, 2, 3], [4, 5, 6]])\n",
    "print(a)\n",
    "print(type(a))\n",
    "\n",
    "# Notice the lack of () - these are properties, not methods\n",
    "print(a.ndim, a.shape, a.dtype) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Creating Arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0., 0., 0.],\n",
       "       [0., 0., 0.],\n",
       "       [0., 0., 0.]])"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a 3x3 array of float zeros\n",
    "a = np.zeros((3, 3))  \n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[1, 1, 1],\n",
       "       [1, 1, 1],\n",
       "       [1, 1, 1],\n",
       "       [1, 1, 1]], dtype=int16)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# NumPy is implemented in C and supports more numerical types than Python\n",
    "# See: https://docs.scipy.org/doc/numpy/user/basics.types.html\n",
    "\n",
    "# Create a 4x3 array of integer ones where integer type (-32768 to 32767)\n",
    "a = np.ones((4, 3), dtype = np.int16)  \n",
    "a"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0.83531804, 0.86343136, 0.9658808 , 0.33879326],\n",
       "       [0.512375  , 0.62835943, 0.39683852, 0.56393082]])"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create a 2x4 array of floats drawn uniformly from [0., 1.)\n",
    "r = np.random.random((2, 4))  \n",
    "r"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Reshaping and Transposing Arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[0 1 2 3 4 5 6 7 8 9]\n",
      "[[0 1]\n",
      " [2 3]\n",
      " [4 5]\n",
      " [6 7]\n",
      " [8 9]]\n"
     ]
    }
   ],
   "source": [
    "# Creates an array of numbers from 0 to input (like Python range() but step could be non-integer)\n",
    "a = np.arange(10)  \n",
    "print(a)\n",
    "a = a.reshape(5, 2)  # Reshape to 5x2\n",
    "print(a)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[0, 2, 4, 6, 8],\n",
       "       [1, 3, 5, 7, 9]])"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "at = np.transpose(a)   # Reverse dimensions of a\n",
    "at"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Indexing Arrays"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "3\n",
      "[4 5 6]\n",
      "[1 2 6]\n"
     ]
    }
   ],
   "source": [
    "a = np.array([[1, 2, 3], [4, 5, 6]])\n",
    "print(a[0, -1])   # First row, last column\n",
    "print(a[1, :])   # Entire second row \n",
    "c = np.array([[True, True, False], [False, False, True]])\n",
    "print(a[c])   # Elements of a that are True in c "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Mathematical Operations with Arrays\n",
    "\n",
    "* Arithmetic operations are element-wise"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[ 10  40  90]\n",
      " [160 250 360]]\n",
      "[[11 22 33]\n",
      " [44 55 66]]\n",
      "[1, 2, 3, 10, 20, 30]\n"
     ]
    }
   ],
   "source": [
    "b = np.array([[10, 20, 30], [40, 50, 60]])\n",
    "print(a * b)\n",
    "print(a + b)  # Arrays must be same shape\n",
    "\n",
    "print([1, 2, 3] + [10, 20, 30])  # In comparison, lists get concatenated"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[  2.71828183   7.3890561   20.08553692]\n",
      " [ 54.59815003 148.4131591  403.42879349]]\n",
      "[[3.16227766 4.47213595 5.47722558]\n",
      " [6.32455532 7.07106781 7.74596669]]\n"
     ]
    }
   ],
   "source": [
    "print(np.exp(a))\n",
    "print(np.sqrt(b))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "* Logical operations return a Boolean array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[False,  True,  True],\n",
       "       [ True,  True,  True]])"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b > 15"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "* Matrix multiplication with `@`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[140, 320],\n",
       "       [320, 770]])"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a @ np.transpose(b)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Working with DataFrames: `pandas`\n",
    "\n",
    "![Pandas](figs/pandas.png \"Pandas\")\n",
    "\n",
    "* Python library for data manipulation and analysis \n",
    "    * The name is derived from **pan**el **da**ta = **pandas**\n",
    "    * Partially written in C, so optimized for performance\n",
    "* Key features\n",
    "    * DataFrame object to hold and manipulate data\n",
    "    * Read from and write in multiple file formats\n",
    "    * Fancy indexing (including hierarchical)\n",
    "    * Operations to handle missing data\n",
    "    * Operations to slice, subset, split, reshape, pivot, merge, and join data sets"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## `DataFrame`\n",
    "\n",
    "* Primary `pandas` object\n",
    "* 2-d heterogeneous tabular data structure with axes labels (rows and columns)\n",
    "* like a `dict` for lists (more precisely, `pd.Series`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Col1</th>\n",
       "      <th>Col2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>b</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Col1 Col2\n",
       "0     1    a\n",
       "1     2    b\n",
       "2     3  NaN"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Create DataFrame from dictionary\n",
    "d = {'Col1': [1, 2, 3], 'Col2': ['a', 'b', np.nan]}\n",
    "df1 = pd.DataFrame(data=d)\n",
    "df1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "      Col1 Col2\n",
      "Row1     1    a\n",
      "Row2     2    b\n",
      "Row3     3  NaN\n"
     ]
    }
   ],
   "source": [
    "# Create DataFrame from list or array\n",
    "df2 = pd.DataFrame(data=[[1, 'a'], [2, 'b'], [3, np.nan]], \n",
    "                  index=['Row1', 'Row2', 'Row3'], columns=['Col1', 'Col2'])\n",
    "print(df2)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Importing and Viewing Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  state region    pop  SATV  SATM  percent  dollars  pay\n",
      "0    AL    ESC   4041   470   514        8    3.648   27\n",
      "1    AK    PAC    550   438   476       42    7.887   43\n",
      "2    AZ    MTN   3665   445   497       25    4.231   30\n",
      "3    AR    WSC   2351   470   511        6    3.334   23\n",
      "4    CA    PAC  29760   419   484       45    4.826   39\n"
     ]
    }
   ],
   "source": [
    "# Data source: https://vincentarelbundock.github.io/Rdatasets/doc/carData/States.html\n",
    "\n",
    "df3 = pd.read_csv('../../data/States.csv')\n",
    "print(df3.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['state', 'region', 'pop', 'SATV', 'SATM', 'percent', 'dollars', 'pay'], dtype='object')\n",
      "state       object\n",
      "region      object\n",
      "pop          int64\n",
      "SATV         int64\n",
      "SATM         int64\n",
      "percent      int64\n",
      "dollars    float64\n",
      "pay          int64\n",
      "dtype: object\n",
      "                pop        SATV        SATM    percent    dollars        pay\n",
      "count     51.000000   51.000000   51.000000  51.000000  51.000000  51.000000\n",
      "mean    4876.647059  448.156863  497.392157  33.745098   5.175490  30.941176\n",
      "std     5439.202691   30.821014   34.568817  24.073922   1.376166   5.308151\n",
      "min      454.000000  397.000000  437.000000   4.000000   2.993000  22.000000\n",
      "25%     1215.000000  422.500000  470.000000  11.500000   4.354000  27.500000\n",
      "50%     3294.000000  443.000000  490.000000  25.000000   5.045000  30.000000\n",
      "75%     5780.000000  474.500000  522.500000  57.500000   5.689500  33.500000\n",
      "max    29760.000000  511.000000  577.000000  74.000000   9.159000  43.000000\n"
     ]
    }
   ],
   "source": [
    "print(df3.columns)\n",
    "print(df3.dtypes)\n",
    "print(df3.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Selecting Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     AL\n",
       "1     AK\n",
       "2     AZ\n",
       "3     AR\n",
       "4     CA\n",
       "5     CO\n",
       "6     CN\n",
       "7     DE\n",
       "8     DC\n",
       "9     FL\n",
       "10    GA\n",
       "11    HI\n",
       "12    ID\n",
       "13    IL\n",
       "14    IN\n",
       "15    IA\n",
       "16    KS\n",
       "17    KY\n",
       "18    LA\n",
       "19    ME\n",
       "20    MD\n",
       "21    MA\n",
       "22    MI\n",
       "23    MN\n",
       "24    MS\n",
       "25    MO\n",
       "26    MT\n",
       "27    NE\n",
       "28    NV\n",
       "29    NH\n",
       "30    NJ\n",
       "31    NM\n",
       "32    NY\n",
       "33    NC\n",
       "34    ND\n",
       "35    OH\n",
       "36    OK\n",
       "37    OR\n",
       "38    PA\n",
       "39    RI\n",
       "40    SC\n",
       "41    SD\n",
       "42    TN\n",
       "43    TX\n",
       "44    UT\n",
       "45    VT\n",
       "46    VA\n",
       "47    WA\n",
       "48    WV\n",
       "49    WI\n",
       "50    WY\n",
       "Name: state, dtype: object"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3.state  # Select column; equivalent to df3['state']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th>region</th>\n",
       "      <th>pop</th>\n",
       "      <th>SATV</th>\n",
       "      <th>SATM</th>\n",
       "      <th>percent</th>\n",
       "      <th>dollars</th>\n",
       "      <th>pay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AL</td>\n",
       "      <td>ESC</td>\n",
       "      <td>4041</td>\n",
       "      <td>470</td>\n",
       "      <td>514</td>\n",
       "      <td>8</td>\n",
       "      <td>3.648</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AK</td>\n",
       "      <td>PAC</td>\n",
       "      <td>550</td>\n",
       "      <td>438</td>\n",
       "      <td>476</td>\n",
       "      <td>42</td>\n",
       "      <td>7.887</td>\n",
       "      <td>43</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AZ</td>\n",
       "      <td>MTN</td>\n",
       "      <td>3665</td>\n",
       "      <td>445</td>\n",
       "      <td>497</td>\n",
       "      <td>25</td>\n",
       "      <td>4.231</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  state region   pop  SATV  SATM  percent  dollars  pay\n",
       "0    AL    ESC  4041   470   514        8    3.648   27\n",
       "1    AK    PAC   550   438   476       42    7.887   43\n",
       "2    AZ    MTN  3665   445   497       25    4.231   30"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3[:3] # Select data by rows"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th>region</th>\n",
       "      <th>pop</th>\n",
       "      <th>SATV</th>\n",
       "      <th>SATM</th>\n",
       "      <th>percent</th>\n",
       "      <th>dollars</th>\n",
       "      <th>pay</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA</td>\n",
       "      <td>PAC</td>\n",
       "      <td>29760</td>\n",
       "      <td>419</td>\n",
       "      <td>484</td>\n",
       "      <td>45</td>\n",
       "      <td>4.826</td>\n",
       "      <td>39</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>FL</td>\n",
       "      <td>SA</td>\n",
       "      <td>12938</td>\n",
       "      <td>418</td>\n",
       "      <td>466</td>\n",
       "      <td>44</td>\n",
       "      <td>5.154</td>\n",
       "      <td>30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>IL</td>\n",
       "      <td>ENC</td>\n",
       "      <td>11431</td>\n",
       "      <td>466</td>\n",
       "      <td>528</td>\n",
       "      <td>16</td>\n",
       "      <td>5.062</td>\n",
       "      <td>34</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>NY</td>\n",
       "      <td>MA</td>\n",
       "      <td>17990</td>\n",
       "      <td>412</td>\n",
       "      <td>470</td>\n",
       "      <td>70</td>\n",
       "      <td>8.500</td>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>OH</td>\n",
       "      <td>ENC</td>\n",
       "      <td>10847</td>\n",
       "      <td>450</td>\n",
       "      <td>499</td>\n",
       "      <td>22</td>\n",
       "      <td>5.639</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>PA</td>\n",
       "      <td>MA</td>\n",
       "      <td>11882</td>\n",
       "      <td>420</td>\n",
       "      <td>463</td>\n",
       "      <td>64</td>\n",
       "      <td>6.534</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>TX</td>\n",
       "      <td>WSC</td>\n",
       "      <td>16987</td>\n",
       "      <td>413</td>\n",
       "      <td>461</td>\n",
       "      <td>42</td>\n",
       "      <td>4.238</td>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   state region    pop  SATV  SATM  percent  dollars  pay\n",
       "4     CA    PAC  29760   419   484       45    4.826   39\n",
       "9     FL     SA  12938   418   466       44    5.154   30\n",
       "13    IL    ENC  11431   466   528       16    5.062   34\n",
       "32    NY     MA  17990   412   470       70    8.500   42\n",
       "35    OH    ENC  10847   450   499       22    5.639   32\n",
       "38    PA     MA  11882   420   463       64    6.534   36\n",
       "43    TX    WSC  16987   413   461       42    4.238   28"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3[df3['pop'] > 10000]  # Select data by column value"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Grouping"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>state</th>\n",
       "      <th>region</th>\n",
       "      <th>pop</th>\n",
       "      <th>SATV</th>\n",
       "      <th>SATM</th>\n",
       "      <th>percent</th>\n",
       "      <th>dollars</th>\n",
       "      <th>pay</th>\n",
       "      <th>SAT</th>\n",
       "      <th>above-avg</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>AL</td>\n",
       "      <td>ESC</td>\n",
       "      <td>4041</td>\n",
       "      <td>470</td>\n",
       "      <td>514</td>\n",
       "      <td>8</td>\n",
       "      <td>3.648</td>\n",
       "      <td>27</td>\n",
       "      <td>984</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>AK</td>\n",
       "      <td>PAC</td>\n",
       "      <td>550</td>\n",
       "      <td>438</td>\n",
       "      <td>476</td>\n",
       "      <td>42</td>\n",
       "      <td>7.887</td>\n",
       "      <td>43</td>\n",
       "      <td>914</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>AZ</td>\n",
       "      <td>MTN</td>\n",
       "      <td>3665</td>\n",
       "      <td>445</td>\n",
       "      <td>497</td>\n",
       "      <td>25</td>\n",
       "      <td>4.231</td>\n",
       "      <td>30</td>\n",
       "      <td>942</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>AR</td>\n",
       "      <td>WSC</td>\n",
       "      <td>2351</td>\n",
       "      <td>470</td>\n",
       "      <td>511</td>\n",
       "      <td>6</td>\n",
       "      <td>3.334</td>\n",
       "      <td>23</td>\n",
       "      <td>981</td>\n",
       "      <td>True</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>CA</td>\n",
       "      <td>PAC</td>\n",
       "      <td>29760</td>\n",
       "      <td>419</td>\n",
       "      <td>484</td>\n",
       "      <td>45</td>\n",
       "      <td>4.826</td>\n",
       "      <td>39</td>\n",
       "      <td>903</td>\n",
       "      <td>False</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  state region    pop  SATV  SATM  percent  dollars  pay  SAT  above-avg\n",
       "0    AL    ESC   4041   470   514        8    3.648   27  984       True\n",
       "1    AK    PAC    550   438   476       42    7.887   43  914      False\n",
       "2    AZ    MTN   3665   445   497       25    4.231   30  942       True\n",
       "3    AR    WSC   2351   470   511        6    3.334   23  981       True\n",
       "4    CA    PAC  29760   419   484       45    4.826   39  903      False"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add a column that indicates whether state performed above median on SAT\n",
    "df3['SAT'] = df3['SATV'] + df3['SATM']\n",
    "df3['above-avg'] = df3['SAT'] > df3['SAT'].median()\n",
    "df3.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>pop</th>\n",
       "      <th>SATV</th>\n",
       "      <th>SATM</th>\n",
       "      <th>percent</th>\n",
       "      <th>dollars</th>\n",
       "      <th>pay</th>\n",
       "      <th>SAT</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>above-avg</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>False</th>\n",
       "      <td>6047.461538</td>\n",
       "      <td>422.076923</td>\n",
       "      <td>468.423077</td>\n",
       "      <td>54.615385</td>\n",
       "      <td>5.940538</td>\n",
       "      <td>33.807692</td>\n",
       "      <td>890.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>True</th>\n",
       "      <td>3659.000000</td>\n",
       "      <td>475.280000</td>\n",
       "      <td>527.520000</td>\n",
       "      <td>12.040000</td>\n",
       "      <td>4.379840</td>\n",
       "      <td>27.960000</td>\n",
       "      <td>1002.8</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   pop        SATV        SATM    percent   dollars  \\\n",
       "above-avg                                                             \n",
       "False      6047.461538  422.076923  468.423077  54.615385  5.940538   \n",
       "True       3659.000000  475.280000  527.520000  12.040000  4.379840   \n",
       "\n",
       "                 pay     SAT  \n",
       "above-avg                     \n",
       "False      33.807692   890.5  \n",
       "True       27.960000  1002.8  "
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Get mean indicators by SAT performance\n",
    "df3.groupby('above-avg').mean(numeric_only=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Estimating Statistical Models: `statsmodels`\n",
    "\n",
    "![Statsmodels](figs/statsmodels_hybi_banner.png \"Statsmodels\")\n",
    "\n",
    "* Python library for estimating statistical models and conducting statistical tests\n",
    "* Uses `numpy` for estimation\n",
    "* Integrates with `pandas` for data handling\n",
    "* Provides an `R`-like formula interface"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Typical Steps\n",
    "\n",
    "1. Get data (`pandas`)\n",
    "* Inspect data (`pandas`, `matplotlib`)\n",
    "* Define and fit model (`statsmodels`)\n",
    "* Inspect model results (`statsmodels`, `matplotlib`)\n",
    "* Inspect model fit (`statsmodels`, `matplotlib`)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## OLS Example: Inspect Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "  state region    pop  SATV  SATM  percent  dollars  pay\n",
      "0    AL    ESC   4041   470   514        8    3.648   27\n",
      "1    AK    PAC    550   438   476       42    7.887   43\n",
      "2    AZ    MTN   3665   445   497       25    4.231   30\n",
      "3    AR    WSC   2351   470   511        6    3.334   23\n",
      "4    CA    PAC  29760   419   484       45    4.826   39\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import statsmodels.api as sm\n",
    "import statsmodels.formula.api as smf\n",
    "\n",
    "# Get data\n",
    "df = pd.read_csv('../../data/States.csv')\n",
    "print(df.head())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                pop        SATV        SATM    percent    dollars        pay\n",
      "count     51.000000   51.000000   51.000000  51.000000  51.000000  51.000000\n",
      "mean    4876.647059  448.156863  497.392157  33.745098   5.175490  30.941176\n",
      "std     5439.202691   30.821014   34.568817  24.073922   1.376166   5.308151\n",
      "min      454.000000  397.000000  437.000000   4.000000   2.993000  22.000000\n",
      "25%     1215.000000  422.500000  470.000000  11.500000   4.354000  27.500000\n",
      "50%     3294.000000  443.000000  490.000000  25.000000   5.045000  30.000000\n",
      "75%     5780.000000  474.500000  522.500000  57.500000   5.689500  33.500000\n",
      "max    29760.000000  511.000000  577.000000  74.000000   9.159000  43.000000\n"
     ]
    }
   ],
   "source": [
    "# Inspect data\n",
    "print(df.describe())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## OLS Example: Fit Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {
    "scrolled": false,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                            OLS Regression Results                            \n",
      "==============================================================================\n",
      "Dep. Variable:                   SATV   R-squared:                       0.782\n",
      "Model:                            OLS   Adj. R-squared:                  0.763\n",
      "Method:                 Least Squares   F-statistic:                     41.35\n",
      "Date:                Mon, 09 Dec 2024   Prob (F-statistic):           1.11e-14\n",
      "Time:                        10:25:06   Log-Likelihood:                -207.81\n",
      "No. Observations:                  51   AIC:                             425.6\n",
      "Df Residuals:                      46   BIC:                             435.3\n",
      "Df Model:                           4                                         \n",
      "Covariance Type:            nonrobust                                         \n",
      "===============================================================================\n",
      "                  coef    std err          t      P>|t|      [0.025      0.975]\n",
      "-------------------------------------------------------------------------------\n",
      "Intercept     508.6092     19.627     25.914      0.000     469.102     548.116\n",
      "percent        -1.2205      0.127     -9.592      0.000      -1.477      -0.964\n",
      "dollars         5.1671      3.320      1.556      0.126      -1.515      11.850\n",
      "pay            -0.4942      0.832     -0.594      0.555      -2.168       1.180\n",
      "np.log(pop)    -3.8421      2.261     -1.699      0.096      -8.393       0.709\n",
      "==============================================================================\n",
      "Omnibus:                        0.275   Durbin-Watson:                   2.137\n",
      "Prob(Omnibus):                  0.872   Jarque-Bera (JB):                0.464\n",
      "Skew:                          -0.061   Prob(JB):                        0.793\n",
      "Kurtosis:                       2.549   Cond. No.                         479.\n",
      "==============================================================================\n",
      "\n",
      "Notes:\n",
      "[1] Standard Errors assume that the covariance matrix of the errors is correctly specified.\n"
     ]
    }
   ],
   "source": [
    "# Define regression model \n",
    "model = smf.ols('SATV ~ percent + dollars + pay + np.log(pop)', data=df)\n",
    "# Fit model\n",
    "results = model.fit()\n",
    "# Inspect results\n",
    "print(results.summary())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Analyzing Networks: `networkx`\n",
    "\n",
    "![NetworkX](figs/networkx.png \"NetworkX\")\n",
    "\n",
    "* Library for studying networks\n",
    "* The major data structures are of the type \"dictionary of dictionaries\"\n",
    "* Capabilities:\n",
    "    * Estimate common network measures\n",
    "    * Construct random networks\n",
    "    * Visualize networks\n",
    "    * Convert networks to and from different formats"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Creating Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[1, 2, 3]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "NodeView((1, 2, 3))"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import networkx as nx\n",
    "\n",
    "# Create an empty network\n",
    "G = nx.Graph()  \n",
    "# Add nodes to network\n",
    "G.add_node(1)\n",
    "G.add_nodes_from([2,3])\n",
    "print(G.nodes())\n",
    "G.nodes()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[(1, 2), (1, 3), (1, 4), (2, 3)]\n",
      "[1, 2, 3, 4]\n"
     ]
    }
   ],
   "source": [
    "# Add edges to network (nodes are automatically added if they don't exist already)\n",
    "G.add_edge(1, 2)\n",
    "G.add_edges_from([(2, 3), (1, 3), (1, 4)])\n",
    "print(G.edges())\n",
    "print(G.nodes())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Node and Edge Attributes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {
    "scrolled": true,
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "NodeDataView({1: {'name': 'Anna'}, 2: {}, 3: {}, 4: {}, 5: {'name': 'Elliot'}, 6: {'name': 'Fathima'}, 7: {'name': 'Fathima'}})"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add/modify node attributes\n",
    "G.nodes[1]['name'] = 'Anna'\n",
    "G.add_node(5, name='Elliot')\n",
    "G.add_nodes_from([6, 7], name='Fathima')\n",
    "G.nodes.data()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "EdgeDataView([(1, 2, {'weight': 1.7}), (1, 3, {'weight': 1}), (1, 4, {}), (2, 3, {'weight': 1.5}), (5, 6, {'weight': 2})])"
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Add/modify edge attributes\n",
    "G.add_edge(5, 6, weight=2)\n",
    "G.add_edges_from([(1, 2), (1, 3)], weight=1)\n",
    "G[1][2]['weight'] = 1.7\n",
    "G.edges[2, 3]['weight'] = 1.5\n",
    "G.edges.data()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Analyzing Networks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DegreeView({1: 3, 2: 2, 3: 2, 4: 1, 5: 1, 6: 1, 7: 0})"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Estimate node degrees -- returns a DegreeView object capable of iterating (node, degree) pairs\n",
    "G.degree()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{1: 0.3333333333333333, 2: 1.0, 3: 1.0, 4: 0, 5: 0, 6: 0, 7: 0}"
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Estimate node clustering\n",
    "# The method returns dictionary -- it's good example about the importance of consistency!\n",
    "nx.clustering(G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<generator object connected_components at 0x7fb942a146d0>"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Identify the connected subcomponents in the network -- returns a generator!\n",
    "nx.connected_components(G)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "fragment"
    }
   },
   "source": [
    "* Despite these inconsistencies, `networkx` is a powerful library for network analysis (although not so much for visualization)\n",
    "* When in doubt or confused, simply consult the documentation!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Machine Learning: `scikit-learn`\n",
    "\n",
    "![Scikit-learn](figs/scikitlearn.png \"Scikit-learn\")\n",
    "\n",
    "* Machine-learning library for Python\n",
    "* Includes various algorithms\n",
    "    * Supervised learning\n",
    "        * Classification\n",
    "        * Regression\n",
    "    * Unsupervised learning\n",
    "        * Clustering\n",
    "        * Dimensionality reduction\n",
    "* Includes classes for model evaluation and model selection\n",
    "* Integrates with `numpy`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## The `scikit-learn` Estimators\n",
    "\n",
    "The estimator is an object (instance of a class) that learns from data. \n",
    "\n",
    "Unsupervised learning estimators have the `fit` method:\n",
    "* `fit(data)`\n",
    "\n",
    "Supervised learning estimators have `fit` and `predict` methods:\n",
    "* `fit(X, y)` to known data\n",
    "* `predict(T)` unknown data\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Plotting: `matplotlib`\n",
    "\n",
    "![Matplotlib](figs/matplotlib.png \"Matplotlib\")\n",
    "\n",
    "* The most used Python library for 2D graphics \n",
    "* To quickly and easily generate figures, use the procedural interface provided by pyplot \n",
    "    * **`pyplot`** provides functions that make `matplotlib` work like MATLAB \n",
    "* To achieve more customization and control, use the object-oriented approach with the **`Axes`** class\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Typical Steps for Plotting with `pyplot`\n",
    "\n",
    "### 1. Import `pyplot` \n",
    "* **`import matplotlib.pyplot as plt`**\n",
    "\n",
    "\\*Set global parameters\n",
    "* `import matplotlib as mpl`\n",
    "* `mpl.rcParams['font.family'] = 'sans-serif'`\n",
    "    \n",
    "\\*Define figure and subplots\n",
    "* `plt.figure(*figsize)`\n",
    "* `plt.subplot(nrow, ncol, plotnum)`\n",
    "    \n",
    "### 2. Plot with plot parameters\n",
    "* **`plt.plot(x, y, *c, *lw, *label)`**\n",
    "* `plt.scatter(x, y, *s, *c, *marker, *alpha, *label)`\n",
    "* `plt.boxplot(x, *positions, *widths)`\n",
    "* `plt.hist(x, *bins, *color)`\n",
    "* `plt.bar(x, y, *width, *color, *yerr)`\n",
    "\n",
    "\\*Adjust specific figure parameters\n",
    "* `plt.xlim(xmin, xmax)`\n",
    "* `plt.xticks(locs, *labels)`\n",
    "\n",
    "\\*Add annotations    \n",
    "* `plt.xlabel(s)`    \n",
    "* `plt.title(s)`\n",
    "* `plt.legend(*loc, *ncol)`\n",
    "* `plt.text(x, y, s, *fontsize)`\n",
    "    \n",
    "### 3. Show or save figure\n",
    "* **`plt.show()`**\n",
    "* `plt.savefig(fname)`\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Simple Plotting"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 640x480 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x = [i for i in range(11)]\n",
    "y = [i**2 for i in x]\n",
    "\n",
    "plt.plot(x, y, 'r--')  # 'r--' is an optional format string indicating color and line type\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "slideshow": {
     "slide_type": "slide"
    }
   },
   "source": [
    "## Making Plots Prettier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {
    "slideshow": {
     "slide_type": "-"
    }
   },
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x300 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "\n",
    "x = [i for i in range(11)]\n",
    "y1 = [i**2 for i in x]\n",
    "y2 = [i**0.5 for i in x]\n",
    "\n",
    "fig = plt.figure(figsize=(8, 3))\n",
    "fig.subplots_adjust(wspace=0.4)\n",
    "\n",
    "plt.subplot(1, 2, 1)\n",
    "plt.plot(x, y1, c='r', lw=0.7, label='y = 2x')\n",
    "plt.legend()\n",
    "plt.xlabel('X')\n",
    "plt.ylabel('Y')\n",
    "plt.title('plot()')\n",
    "\n",
    "plt.subplot(1, 2, 2)\n",
    "plt.scatter(x, y2, c='yellowgreen', s=10, label='y = x^2')\n",
    "plt.legend()\n",
    "plt.xlabel('X')\n",
    "plt.ylabel('Y')\n",
    "plt.title('scatter()')\n",
    "\n",
    "plt.savefig('figs/pyplot_demo.pdf')\n",
    "plt.show()"
   ]
  }
 ],
 "metadata": {
  "celltoolbar": "Slideshow",
  "kernelspec": {
   "display_name": "base",
   "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": "3.8.19"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}