01-numpy.ipynb

   1 {
   2  "metadata": {
   3   "name": ""
   4  },
   5  "nbformat": 3,
   6  "nbformat_minor": 0,
   7  "worksheets": [
   8   {
   9    "cells": [
  10     {
  11      "cell_type": "heading",
  12      "level": 2,
  13      "metadata": {
  14       "cell_tags": []
  15      },
  16      "source": [
  17       "Analyzing Patient Data"
  18      ]
  19     },
  20     {
  21      "cell_type": "markdown",
  22      "metadata": {
  23       "cell_tags": []
  24      },
  25      "source": [
  26       "We are studying inflammation in patients who have been given a new treatment for arthritis,\n",
  27       "and need to analyze the first dozen data sets.\n",
  28       "The data sets are stored in [comma-separated values](../../gloss.html#csv) (CSV) format:\n",
  29       "each row holds information for a single patient,\n",
  30       "and the columns represent successive days.\n",
  31       "The first few rows of our first file look like this:\n",
  32       "\n",
  33       "    0,0,1,3,1,2,4,7,8,3,3,3,10,5,7,4,7,7,12,18,6,13,11,11,7,7,4,6,8,8,4,4,5,7,3,4,2,3,0,0\n",
  34       "    0,1,2,1,2,1,3,2,2,6,10,11,5,9,4,4,7,16,8,6,18,4,12,5,12,7,11,5,11,3,3,5,4,4,5,5,1,1,0,1\n",
  35       "    0,1,1,3,3,2,6,2,5,9,5,7,4,5,4,15,5,11,9,10,19,14,12,17,7,12,11,7,4,2,10,5,4,2,2,3,2,2,1,1\n",
  36       "    0,0,2,0,4,2,2,1,6,7,10,7,9,13,8,8,15,10,10,7,17,4,4,7,6,15,6,4,9,11,3,5,6,3,3,4,2,3,2,1\n",
  37       "    0,1,1,3,3,1,3,5,2,4,4,7,6,5,3,10,8,10,6,17,9,14,9,7,13,9,12,6,7,7,9,6,3,2,2,4,2,0,1,1"
  38      ]
  39     },
  40     {
  41      "cell_type": "markdown",
  42      "metadata": {
  43       "cell_tags": []
  44      },
  45      "source": [
  46       "We want to:\n",
  47       "\n",
  48       "*   load that data into memory,\n",
  49       "*   calculate the average inflammation per day across all patients, and\n",
  50       "*   plot the result.\n",
  51       "\n",
  52       "To do all that, we'll have to learn a little bit about programming."
  53      ]
  54     },
  55     {
  56      "cell_type": "markdown",
  57      "metadata": {
  58       "cell_tags": [
  59        "objectives"
  60       ]
  61      },
  62      "source": [
  63       "#### Objectives\n",
  64       "\n",
  65       "*   Explain what a library is, and what libraries are used for.\n",
  66       "*   Load a Python library and use the things it contains.\n",
  67       "*   Read tabular data from a file into a program.\n",
  68       "*   Assign values to variables.\n",
  69       "*   Select individual values and subsections from data.\n",
  70       "*   Perform operations on arrays of data.\n",
  71       "*   Display simple graphs."
  72      ]
  73     },
  74     {
  75      "cell_type": "heading",
  76      "level": 3,
  77      "metadata": {
  78       "cell_tags": []
  79      },
  80      "source": [
  81       "Loading Data"
  82      ]
  83     },
  84     {
  85      "cell_type": "markdown",
  86      "metadata": {
  87       "cell_tags": []
  88      },
  89      "source": [
  90       "Words are useful,\n",
  91       "but what's more useful are the sentences and stories we use them to build.\n",
  92       "Similarly,\n",
  93       "while a lot of powerful tools are built into languages like Python,\n",
  94       "even more lives in the [libraries](../../gloss.html#library) they are used to build.\n",
  95       "\n",
  96       "In order to load our inflammation data,\n",
  97       "we need to [import](../../gloss.html#import) a library called NumPy\n",
  98       "that knows how to operate on matrices:"
  99      ]
 100     },
 101     {
 102      "cell_type": "code",
 103      "collapsed": false,
 104      "input": [
 105       "import numpy"
 106      ],
 107      "language": "python",
 108      "metadata": {
 109       "cell_tags": []
 110      },
 111      "outputs": [],
 112      "prompt_number": 1
 113     },
 114     {
 115      "cell_type": "markdown",
 116      "metadata": {
 117       "cell_tags": []
 118      },
 119      "source": [
 120       "Importing a library is like getting a piece of lab equipment out of a storage locker\n",
 121       "and setting it up on the bench.\n",
 122       "Once it's done,\n",
 123       "we can ask the library to read our data file for us:"
 124      ]
 125     },
 126     {
 127      "cell_type": "code",
 128      "collapsed": false,
 129      "input": [
 130       "numpy.loadtxt(fname='inflammation-01.csv', delimiter=',')"
 131      ],
 132      "language": "python",
 133      "metadata": {
 134       "cell_tags": []
 135      },
 136      "outputs": [
 137       {
 138        "metadata": {},
 139        "output_type": "pyout",
 140        "prompt_number": 2,
 141        "text": [
 142         "array([[ 0.,  0.,  1., ...,  3.,  0.,  0.],\n",
 143         "       [ 0.,  1.,  2., ...,  1.,  0.,  1.],\n",
 144         "       [ 0.,  1.,  1., ...,  2.,  1.,  1.],\n",
 145         "       ..., \n",
 146         "       [ 0.,  1.,  1., ...,  1.,  1.,  1.],\n",
 147         "       [ 0.,  0.,  0., ...,  0.,  2.,  0.],\n",
 148         "       [ 0.,  0.,  1., ...,  1.,  1.,  0.]])"
 149        ]
 150       }
 151      ],
 152      "prompt_number": 2
 153     },
 154     {
 155      "cell_type": "markdown",
 156      "metadata": {
 157       "cell_tags": []
 158      },
 159      "source": [
 160       "The expression `numpy.loadtxt(...)` is a [function call](../../gloss.html#function-call)\n",
 161       "that asks Python to run the function `loadtxt` that belongs to the `numpy` library.\n",
 162       "This [dotted notation](../../gloss.html#dotted-notation) is used everywhere in Python\n",
 163       "to refer to the parts of things as `whole.part`.\n",
 164       "\n",
 165       "`numpy.loadtxt` has two [parameters](../../gloss.html#parameter):\n",
 166       "the name of the file we want to read,\n",
 167       "and the [delimiter](../../gloss.html#delimiter) that separates values on a line.\n",
 168       "These both need to be character strings (or [strings](../../gloss.html#string) for short),\n",
 169       "so we put them in quotes.\n",
 170       "\n",
 171       "When we are finished typing and press Shift+Enter,\n",
 172       "the notebook runs our command.\n",
 173       "Since we haven't told it to do anything else with the function's output,\n",
 174       "the notebook displays it.\n",
 175       "In this case,\n",
 176       "that output is the data we just loaded.\n",
 177       "By default,\n",
 178       "only a few rows and columns are shown\n",
 179       "(with `...` displayed to mark missing data).\n",
 180       "To save space,\n",
 181       "Python displays numbers as `1.` instead of `1.0`\n",
 182       "when there's nothing interesting after the decimal point."
 183      ]
 184     },
 185     {
 186      "cell_type": "markdown",
 187      "metadata": {
 188       "cell_tags": []
 189      },
 190      "source": [
 191       "Our call to `numpy.loadtxt` read our file,\n",
 192       "but didn't save the data in memory.\n",
 193       "To do that,\n",
 194       "we need to [assign](../../gloss.html#assignment) the array to a [variable](../../gloss.html#variable).\n",
 195       "A variable is just a name for a value,\n",
 196       "such as `x`, `current_temperature`, or `subject_id`.\n",
 197       "We can create a new variable simply by assigning a value to it using `=`:"
 198      ]
 199     },
 200     {
 201      "cell_type": "code",
 202      "collapsed": false,
 203      "input": [
 204       "weight_kg = 55"
 205      ],
 206      "language": "python",
 207      "metadata": {
 208       "cell_tags": []
 209      },
 210      "outputs": [],
 211      "prompt_number": 3
 212     },
 213     {
 214      "cell_type": "markdown",
 215      "metadata": {
 216       "cell_tags": []
 217      },
 218      "source": [
 219       "Once a variable has a value, we can print it:"
 220      ]
 221     },
 222     {
 223      "cell_type": "code",
 224      "collapsed": false,
 225      "input": [
 226       "print weight_kg"
 227      ],
 228      "language": "python",
 229      "metadata": {
 230       "cell_tags": []
 231      },
 232      "outputs": [
 233       {
 234        "output_type": "stream",
 235        "stream": "stdout",
 236        "text": [
 237         "55\n"
 238        ]
 239       }
 240      ],
 241      "prompt_number": 4
 242     },
 243     {
 244      "cell_type": "markdown",
 245      "metadata": {
 246       "cell_tags": []
 247      },
 248      "source": [
 249       "and do arithmetic with it:"
 250      ]
 251     },
 252     {
 253      "cell_type": "code",
 254      "collapsed": false,
 255      "input": [
 256       "print 'weight in pounds:', 2.2 * weight_kg"
 257      ],
 258      "language": "python",
 259      "metadata": {
 260       "cell_tags": []
 261      },
 262      "outputs": [
 263       {
 264        "output_type": "stream",
 265        "stream": "stdout",
 266        "text": [
 267         "weight in pounds: 121.0\n"
 268        ]
 269       }
 270      ],
 271      "prompt_number": 5
 272     },
 273     {
 274      "cell_type": "markdown",
 275      "metadata": {
 276       "cell_tags": []
 277      },
 278      "source": [
 279       "We can also change a variable's value by assigning it a new one:"
 280      ]
 281     },
 282     {
 283      "cell_type": "code",
 284      "collapsed": false,
 285      "input": [
 286       "weight_kg = 57.5\n",
 287       "print 'weight in kilograms is now:', weight_kg"
 288      ],
 289      "language": "python",
 290      "metadata": {
 291       "cell_tags": []
 292      },
 293      "outputs": [
 294       {
 295        "output_type": "stream",
 296        "stream": "stdout",
 297        "text": [
 298         "weight in kilograms is now: 57.5\n"
 299        ]
 300       }
 301      ],
 302      "prompt_number": 6
 303     },
 304     {
 305      "cell_type": "markdown",
 306      "metadata": {
 307       "cell_tags": []
 308      },
 309      "source": [
 310       "As the example above shows,\n",
 311       "we can print several things at once by separating them with commas.\n",
 312       "\n",
 313       "If we imagine the variable as a sticky note with a name written on it,\n",
 314       "assignment is like putting the sticky note on a particular value:"
 315      ]
 316     },
 317     {
 318      "cell_type": "markdown",
 319      "metadata": {
 320       "cell_tags": []
 321      },
 322      "source": [
 323       "<img src=\"files/img/python-sticky-note-variables-01.svg\" alt=\"Variables as Sticky Notes\" />"
 324      ]
 325     },
 326     {
 327      "cell_type": "markdown",
 328      "metadata": {
 329       "cell_tags": []
 330      },
 331      "source": [
 332       "This means that assigning a value to one variable does *not* change the values of other variables.\n",
 333       "For example,\n",
 334       "let's store the subject's weight in pounds in a variable:"
 335      ]
 336     },
 337     {
 338      "cell_type": "code",
 339      "collapsed": false,
 340      "input": [
 341       "weight_lb = 2.2 * weight_kg\n",
 342       "print 'weight in kilograms:', weight_kg, 'and in pounds:', weight_lb"
 343      ],
 344      "language": "python",
 345      "metadata": {
 346       "cell_tags": []
 347      },
 348      "outputs": [
 349       {
 350        "output_type": "stream",
 351        "stream": "stdout",
 352        "text": [
 353         "weight in kilograms: 57.5 and in pounds: 126.5\n"
 354        ]
 355       }
 356      ],
 357      "prompt_number": 7
 358     },
 359     {
 360      "cell_type": "markdown",
 361      "metadata": {},
 362      "source": [
 363       "<img src=\"files/img/python-sticky-note-variables-02.svg\" alt=\"Creating Another Variable\" />"
 364      ]
 365     },
 366     {
 367      "cell_type": "markdown",
 368      "metadata": {
 369       "cell_tags": []
 370      },
 371      "source": [
 372       "and then change `weight_kg`:"
 373      ]
 374     },
 375     {
 376      "cell_type": "code",
 377      "collapsed": false,
 378      "input": [
 379       "weight_kg = 100.0\n",
 380       "print 'weight in kilograms is now:', weight_kg, 'and weight in pounds is still:', weight_lb"
 381      ],
 382      "language": "python",
 383      "metadata": {
 384       "cell_tags": []
 385      },
 386      "outputs": [
 387       {
 388        "output_type": "stream",
 389        "stream": "stdout",
 390        "text": [
 391         "weight in kilograms is now: 100.0 and weight in pounds is still: 126.5\n"
 392        ]
 393       }
 394      ],
 395      "prompt_number": 8
 396     },
 397     {
 398      "cell_type": "markdown",
 399      "metadata": {},
 400      "source": [
 401       "<img src=\"files/img/python-sticky-note-variables-03.svg\" alt=\"Updating a Variable\" />"
 402      ]
 403     },
 404     {
 405      "cell_type": "markdown",
 406      "metadata": {
 407       "cell_tags": []
 408      },
 409      "source": [
 410       "Since `weight_lb` doesn't \"remember\" where its value came from,\n",
 411       "it isn't automatically updated when `weight_kg` changes.\n",
 412       "This is different from the way spreadsheets work.\n",
 413       "\n",
 414       "Now that we know how to assign things to variables,\n",
 415       "let's re-run `numpy.loadtxt` and save its result:"
 416      ]
 417     },
 418     {
 419      "cell_type": "code",
 420      "collapsed": false,
 421      "input": [
 422       "data = numpy.loadtxt(fname='inflammation-01.csv', delimiter=',')"
 423      ],
 424      "language": "python",
 425      "metadata": {
 426       "cell_tags": []
 427      },
 428      "outputs": [],
 429      "prompt_number": 9
 430     },
 431     {
 432      "cell_type": "markdown",
 433      "metadata": {
 434       "cell_tags": []
 435      },
 436      "source": [
 437       "This statement doesn't produce any output because assignment doesn't display anything.\n",
 438       "If we want to check that our data has been loaded,\n",
 439       "we can print the variable's value:"
 440      ]
 441     },
 442     {
 443      "cell_type": "code",
 444      "collapsed": false,
 445      "input": [
 446       "print data"
 447      ],
 448      "language": "python",
 449      "metadata": {
 450       "cell_tags": []
 451      },
 452      "outputs": [
 453       {
 454        "output_type": "stream",
 455        "stream": "stdout",
 456        "text": [
 457         "[[ 0.  0.  1. ...,  3.  0.  0.]\n",
 458         " [ 0.  1.  2. ...,  1.  0.  1.]\n",
 459         " [ 0.  1.  1. ...,  2.  1.  1.]\n",
 460         " ..., \n",
 461         " [ 0.  1.  1. ...,  1.  1.  1.]\n",
 462         " [ 0.  0.  0. ...,  0.  2.  0.]\n",
 463         " [ 0.  0.  1. ...,  1.  1.  0.]]\n"
 464        ]
 465       }
 466      ],
 467      "prompt_number": 10
 468     },
 469     {
 470      "cell_type": "markdown",
 471      "metadata": {
 472       "cell_tags": [
 473        "challenges"
 474       ]
 475      },
 476      "source": [
 477       "#### Challenges\n",
 478       "\n",
 479       "1.  Draw diagrams showing what variables refer to what values after each statement in the following program:\n",
 480       "\n",
 481       "    ~~~python\n",
 482       "    mass = 47.5\n",
 483       "    age = 122\n",
 484       "    mass = mass * 2.0\n",
 485       "    age = age - 20\n",
 486       "    ~~~\n",
 487       "\n",
 488       "1.  What does the following program print out?\n",
 489       "    ~~~python\n",
 490       "    first, second = 'Grace', 'Hopper'\n",
 491       "    third, fourth = second, first\n",
 492       "    print third, fourth\n",
 493       "    ~~~"
 494      ]
 495     },
 496     {
 497      "cell_type": "heading",
 498      "level": 3,
 499      "metadata": {
 500       "cell_tags": []
 501      },
 502      "source": [
 503       "Manipulating Data"
 504      ]
 505     },
 506     {
 507      "cell_type": "markdown",
 508      "metadata": {
 509       "cell_tags": []
 510      },
 511      "source": [
 512       "Now that our data is in memory,\n",
 513       "we can start doing things with it.\n",
 514       "First,\n",
 515       "let's ask what [type](../../gloss.html#data-type) of thing `data` refers to:"
 516      ]
 517     },
 518     {
 519      "cell_type": "code",
 520      "collapsed": false,
 521      "input": [
 522       "print type(data)"
 523      ],
 524      "language": "python",
 525      "metadata": {
 526       "cell_tags": []
 527      },
 528      "outputs": [
 529       {
 530        "output_type": "stream",
 531        "stream": "stdout",
 532        "text": [
 533         "<type 'numpy.ndarray'>\n"
 534        ]
 535       }
 536      ],
 537      "prompt_number": 11
 538     },
 539     {
 540      "cell_type": "markdown",
 541      "metadata": {
 542       "cell_tags": []
 543      },
 544      "source": [
 545       "The output tells us that `data` currently refers to an N-dimensional array created by the NumPy library.\n",
 546       "We can see what its [shape](../../gloss.html#shape) is like this:"
 547      ]
 548     },
 549     {
 550      "cell_type": "code",
 551      "collapsed": false,
 552      "input": [
 553       "print data.shape"
 554      ],
 555      "language": "python",
 556      "metadata": {
 557       "cell_tags": []
 558      },
 559      "outputs": [
 560       {
 561        "output_type": "stream",
 562        "stream": "stdout",
 563        "text": [
 564         "(60, 40)\n"
 565        ]
 566       }
 567      ],
 568      "prompt_number": 12
 569     },
 570     {
 571      "cell_type": "markdown",
 572      "metadata": {
 573       "cell_tags": []
 574      },
 575      "source": [
 576       "This tells us that `data` has 60 rows and 40 columns.\n",
 577       "`data.shape` is a [member](../../gloss.html#member) of `data`,\n",
 578       "i.e.,\n",
 579       "a value that is stored as part of a larger value.\n",
 580       "We use the same dotted notation for the members of values\n",
 581       "that we use for the functions in libraries\n",
 582       "because they have the same part-and-whole relationship."
 583      ]
 584     },
 585     {
 586      "cell_type": "markdown",
 587      "metadata": {
 588       "cell_tags": []
 589      },
 590      "source": [
 591       "If we want to get a single value from the matrix,\n",
 592       "we must provide an [index](../../gloss.html#index) in square brackets,\n",
 593       "just as we do in math:"
 594      ]
 595     },
 596     {
 597      "cell_type": "code",
 598      "collapsed": false,
 599      "input": [
 600       "print 'first value in data:', data[0, 0]"
 601      ],
 602      "language": "python",
 603      "metadata": {
 604       "cell_tags": []
 605      },
 606      "outputs": [
 607       {
 608        "output_type": "stream",
 609        "stream": "stdout",
 610        "text": [
 611         "first value in data: 0.0\n"
 612        ]
 613       }
 614      ],
 615      "prompt_number": 13
 616     },
 617     {
 618      "cell_type": "code",
 619      "collapsed": false,
 620      "input": [
 621       "print 'middle value in data:', data[30, 20]"
 622      ],
 623      "language": "python",
 624      "metadata": {
 625       "cell_tags": []
 626      },
 627      "outputs": [
 628       {
 629        "output_type": "stream",
 630        "stream": "stdout",
 631        "text": [
 632         "middle value in data: 13.0\n"
 633        ]
 634       }
 635      ],
 636      "prompt_number": 14
 637     },
 638     {
 639      "cell_type": "markdown",
 640      "metadata": {
 641       "cell_tags": []
 642      },
 643      "source": [
 644       "The expression `data[30, 20]` may not surprise you,\n",
 645       "but `data[0, 0]` might.\n",
 646       "Programming languages like Fortran and MATLAB start counting at 1,\n",
 647       "because that's what human beings have done for thousands of years.\n",
 648       "Languages in the C family (including C++, Java, Perl, and Python) count from 0\n",
 649       "because that's simpler for computers to do.\n",
 650       "As a result,\n",
 651       "if we have an M&times;N array in Python,\n",
 652       "its indices go from 0 to M-1 on the first axis\n",
 653       "and 0 to N-1 on the second.\n",
 654       "It takes a bit of getting used to,\n",
 655       "but one way to remember the rule is that\n",
 656       "the index is how many steps we have to take from the start to get the item we want.\n",
 657       "\n",
 658       "> #### In the Corner\n",
 659       ">\n",
 660       "> What may also surprise you is that when Python displays an array,\n",
 661       "> it shows the element with index `[0, 0]` in the upper left corner\n",
 662       "> rather than the lower left.\n",
 663       "> This is consistent with the way mathematicians draw matrices,\n",
 664       "> but different from the Cartesian coordinates.\n",
 665       "> The indices are (row, column) instead of (column, row) for the same reason."
 666      ]
 667     },
 668     {
 669      "cell_type": "markdown",
 670      "metadata": {
 671       "cell_tags": []
 672      },
 673      "source": [
 674       "An index like `[30, 20]` selects a single element of an array,\n",
 675       "but we can select whole sections as well.\n",
 676       "For example,\n",
 677       "we can select the first ten days (columns) of values\n",
 678       "for the first four (rows) patients like this:"
 679      ]
 680     },
 681     {
 682      "cell_type": "code",
 683      "collapsed": false,
 684      "input": [
 685       "print data[0:4, 0:10]"
 686      ],
 687      "language": "python",
 688      "metadata": {
 689       "cell_tags": []
 690      },
 691      "outputs": [
 692       {
 693        "output_type": "stream",
 694        "stream": "stdout",
 695        "text": [
 696         "[[ 0.  0.  1.  3.  1.  2.  4.  7.  8.  3.]\n",
 697         " [ 0.  1.  2.  1.  2.  1.  3.  2.  2.  6.]\n",
 698         " [ 0.  1.  1.  3.  3.  2.  6.  2.  5.  9.]\n",
 699         " [ 0.  0.  2.  0.  4.  2.  2.  1.  6.  7.]]\n"
 700        ]
 701       }
 702      ],
 703      "prompt_number": 15
 704     },
 705     {
 706      "cell_type": "markdown",
 707      "metadata": {
 708       "cell_tags": []
 709      },
 710      "source": [
 711       "The [slice](../../gloss.html#slice) `0:4` means,\n",
 712       "\"Start at index 0 and go up to, but not including, index 4.\"\n",
 713       "Again,\n",
 714       "the up-to-but-not-including takes a bit of getting used to,\n",
 715       "but the rule is that the difference between the upper and lower bounds is the number of values in the slice.\n",
 716       "\n",
 717       "We don't have to start slices at 0:"
 718      ]
 719     },
 720     {
 721      "cell_type": "code",
 722      "collapsed": false,
 723      "input": [
 724       "print data[5:10, 0:10]"
 725      ],
 726      "language": "python",
 727      "metadata": {
 728       "cell_tags": []
 729      },
 730      "outputs": [
 731       {
 732        "output_type": "stream",
 733        "stream": "stdout",
 734        "text": [
 735         "[[ 0.  0.  1.  2.  2.  4.  2.  1.  6.  4.]\n",
 736         " [ 0.  0.  2.  2.  4.  2.  2.  5.  5.  8.]\n",
 737         " [ 0.  0.  1.  2.  3.  1.  2.  3.  5.  3.]\n",
 738         " [ 0.  0.  0.  3.  1.  5.  6.  5.  5.  8.]\n",
 739         " [ 0.  1.  1.  2.  1.  3.  5.  3.  5.  8.]]\n"
 740        ]
 741       }
 742      ],
 743      "prompt_number": 16
 744     },
 745     {
 746      "cell_type": "markdown",
 747      "metadata": {
 748       "cell_tags": []
 749      },
 750      "source": [
 751       "and we don't have to take all the values in the slice---if we provide a [stride](../../gloss.html#stride),\n",
 752       "Python takes values spaced that far apart:"
 753      ]
 754     },
 755     {
 756      "cell_type": "code",
 757      "collapsed": false,
 758      "input": [
 759       "print data[0:10:3, 0:10:2]"
 760      ],
 761      "language": "python",
 762      "metadata": {
 763       "cell_tags": []
 764      },
 765      "outputs": [
 766       {
 767        "output_type": "stream",
 768        "stream": "stdout",
 769        "text": [
 770         "[[ 0.  1.  1.  4.  8.]\n",
 771         " [ 0.  2.  4.  2.  6.]\n",
 772         " [ 0.  2.  4.  2.  5.]\n",
 773         " [ 0.  1.  1.  5.  5.]]\n"
 774        ]
 775       }
 776      ],
 777      "prompt_number": 17
 778     },
 779     {
 780      "cell_type": "markdown",
 781      "metadata": {
 782       "cell_tags": []
 783      },
 784      "source": [
 785       "Here,\n",
 786       "we have taken rows 0, 3, 6, and 9,\n",
 787       "and columns 0, 2, 4, 6, and 8.\n",
 788       "(Again, we always include the lower bound,\n",
 789       "but stop when we reach or cross the upper bound.)"
 790      ]
 791     },
 792     {
 793      "cell_type": "markdown",
 794      "metadata": {
 795       "cell_tags": []
 796      },
 797      "source": [
 798       "We also don't have to include the upper and lower bound on the slice.\n",
 799       "If we don't include the lower bound,\n",
 800       "Python uses 0 by default;\n",
 801       "if we don't include the upper,\n",
 802       "the slice runs to the end of the axis,\n",
 803       "and if we don't include either\n",
 804       "(i.e., if we just use ':' on its own),\n",
 805       "the slice includes everything:"
 806      ]
 807     },
 808     {
 809      "cell_type": "code",
 810      "collapsed": false,
 811      "input": [
 812       "small = data[:3, 36:]\n",
 813       "print 'small is:'\n",
 814       "print small"
 815      ],
 816      "language": "python",
 817      "metadata": {
 818       "cell_tags": []
 819      },
 820      "outputs": [
 821       {
 822        "output_type": "stream",
 823        "stream": "stdout",
 824        "text": [
 825         "small is:\n",
 826         "[[ 2.  3.  0.  0.]\n",
 827         " [ 1.  1.  0.  1.]\n",
 828         " [ 2.  2.  1.  1.]]\n"
 829        ]
 830       }
 831      ],
 832      "prompt_number": 18
 833     },
 834     {
 835      "cell_type": "markdown",
 836      "metadata": {
 837       "cell_tags": []
 838      },
 839      "source": [
 840       "Arrays also know how to perform common mathematical operations on their values.\n",
 841       "If we want to find the average inflammation for all patients on all days,\n",
 842       "for example,\n",
 843       "we can just ask the array for its mean value"
 844      ]
 845     },
 846     {
 847      "cell_type": "code",
 848      "collapsed": false,
 849      "input": [
 850       "print data.mean()"
 851      ],
 852      "language": "python",
 853      "metadata": {
 854       "cell_tags": []
 855      },
 856      "outputs": [
 857       {
 858        "output_type": "stream",
 859        "stream": "stdout",
 860        "text": [
 861         "6.14875\n"
 862        ]
 863       }
 864      ],
 865      "prompt_number": 19
 866     },
 867     {
 868      "cell_type": "markdown",
 869      "metadata": {
 870       "cell_tags": []
 871      },
 872      "source": [
 873       "`mean` is a [method](../../gloss.html#method) of the array,\n",
 874       "i.e.,\n",
 875       "a function that belongs to it\n",
 876       "in the same way that the member `shape` does.\n",
 877       "If variables are nouns, methods are verbs:\n",
 878       "they are what the thing in question knows how to do.\n",
 879       "This is why `data.shape` doesn't need to be called\n",
 880       "(it's just a thing)\n",
 881       "but `data.mean()` does\n",
 882       "(it's an action).\n",
 883       "It is also why we need empty parentheses for `data.mean()`:\n",
 884       "even when we're not passing in any parameters,\n",
 885       "parentheses are how we tell Python to go and do something for us.\n",
 886       "\n",
 887       "NumPy arrays have lots of useful methods:"
 888      ]
 889     },
 890     {
 891      "cell_type": "code",
 892      "collapsed": false,
 893      "input": [
 894       "print 'maximum inflammation:', data.max()\n",
 895       "print 'minimum inflammation:', data.min()\n",
 896       "print 'standard deviation:', data.std()"
 897      ],
 898      "language": "python",
 899      "metadata": {
 900       "cell_tags": []
 901      },
 902      "outputs": [
 903       {
 904        "output_type": "stream",
 905        "stream": "stdout",
 906        "text": [
 907         "maximum inflammation: 20.0\n",
 908         "minimum inflammation: 0.0\n",
 909         "standard deviation: 4.61383319712\n"
 910        ]
 911       }
 912      ],
 913      "prompt_number": 20
 914     },
 915     {
 916      "cell_type": "markdown",
 917      "metadata": {
 918       "cell_tags": []
 919      },
 920      "source": [
 921       "When analyzing data,\n",
 922       "though,\n",
 923       "we often want to look at partial statistics,\n",
 924       "such as the maximum value per patient\n",
 925       "or the average value per day.\n",
 926       "One way to do this is to select the data we want to create a new temporary array,\n",
 927       "then ask it to do the calculation:"
 928      ]
 929     },
 930     {
 931      "cell_type": "code",
 932      "collapsed": false,
 933      "input": [
 934       "patient_0 = data[0, :] # 0 on the first axis, everything on the second\n",
 935       "print 'maximum inflammation for patient 0:', patient_0.max()"
 936      ],
 937      "language": "python",
 938      "metadata": {
 939       "cell_tags": []
 940      },
 941      "outputs": [
 942       {
 943        "output_type": "stream",
 944        "stream": "stdout",
 945        "text": [
 946         "maximum inflammation for patient 0: 18.0\n"
 947        ]
 948       }
 949      ],
 950      "prompt_number": 21
 951     },
 952     {
 953      "cell_type": "markdown",
 954      "metadata": {
 955       "cell_tags": []
 956      },
 957      "source": [
 958       "We don't actually need to store the row in a variable of its own.\n",
 959       "Instead, we can combine the selection and the method call:"
 960      ]
 961     },
 962     {
 963      "cell_type": "code",
 964      "collapsed": false,
 965      "input": [
 966       "print 'maximum inflammation for patient 2:', data[2, :].max()"
 967      ],
 968      "language": "python",
 969      "metadata": {
 970       "cell_tags": []
 971      },
 972      "outputs": [
 973       {
 974        "output_type": "stream",
 975        "stream": "stdout",
 976        "text": [
 977         "maximum inflammation for patient 2: 19.0\n"
 978        ]
 979       }
 980      ],
 981      "prompt_number": 22
 982     },
 983     {
 984      "cell_type": "markdown",
 985      "metadata": {
 986       "cell_tags": []
 987      },
 988      "source": [
 989       "What if we need the maximum inflammation for *all* patients,\n",
 990       "or the average for each day?\n",
 991       "As the diagram below shows,\n",
 992       "we want to perform the operation across an axis:"
 993      ]
 994     },
 995     {
 996      "cell_type": "markdown",
 997      "metadata": {
 998       "cell_tags": []
 999      },
1000      "source": [
1001       "<img src=\"files/img/python-operations-across-axes.svg\" alt=\"Operations Across Axes\" />"
1002      ]
1003     },
1004     {
1005      "cell_type": "markdown",
1006      "metadata": {
1007       "cell_tags": []
1008      },
1009      "source": [
1010       "To support this,\n",
1011       "most array methods allow us to specify the axis we want to work on.\n",
1012       "If we ask for the average across axis 0,\n",
1013       "we get:"
1014      ]
1015     },
1016     {
1017      "cell_type": "code",
1018      "collapsed": false,
1019      "input": [
1020       "print data.mean(axis=0)"
1021      ],
1022      "language": "python",
1023      "metadata": {
1024       "cell_tags": []
1025      },
1026      "outputs": [
1027       {
1028        "output_type": "stream",
1029        "stream": "stdout",
1030        "text": [
1031         "[  0.           0.45         1.11666667   1.75         2.43333333   3.15\n",
1032         "   3.8          3.88333333   5.23333333   5.51666667   5.95         5.9\n",
1033         "   8.35         7.73333333   8.36666667   9.5          9.58333333\n",
1034         "  10.63333333  11.56666667  12.35        13.25        11.96666667\n",
1035         "  11.03333333  10.16666667  10.           8.66666667   9.15         7.25\n",
1036         "   7.33333333   6.58333333   6.06666667   5.95         5.11666667   3.6\n",
1037         "   3.3          3.56666667   2.48333333   1.5          1.13333333\n",
1038         "   0.56666667]\n"
1039        ]
1040       }
1041      ],
1042      "prompt_number": 23
1043     },
1044     {
1045      "cell_type": "markdown",
1046      "metadata": {
1047       "cell_tags": []
1048      },
1049      "source": [
1050       "As a quick check,\n",
1051       "we can ask this array what its shape is:"
1052      ]
1053     },
1054     {
1055      "cell_type": "code",
1056      "collapsed": false,
1057      "input": [
1058       "print data.mean(axis=0).shape"
1059      ],
1060      "language": "python",
1061      "metadata": {
1062       "cell_tags": []
1063      },
1064      "outputs": [
1065       {
1066        "output_type": "stream",
1067        "stream": "stdout",
1068        "text": [
1069         "(40,)\n"
1070        ]
1071       }
1072      ],
1073      "prompt_number": 24
1074     },
1075     {
1076      "cell_type": "markdown",
1077      "metadata": {
1078       "cell_tags": []
1079      },
1080      "source": [
1081       "The expression `(40,)` tells us we have an N&times;1 vector,\n",
1082       "so this is the average inflammation per day for all patients.\n",
1083       "If we average across axis 1, we get:"
1084      ]
1085     },
1086     {
1087      "cell_type": "code",
1088      "collapsed": false,
1089      "input": [
1090       "print data.mean(axis=1)"
1091      ],
1092      "language": "python",
1093      "metadata": {
1094       "cell_tags": []
1095      },
1096      "outputs": [
1097       {
1098        "output_type": "stream",
1099        "stream": "stdout",
1100        "text": [
1101         "[ 5.45   5.425  6.1    5.9    5.55   6.225  5.975  6.65   6.625  6.525\n",
1102         "  6.775  5.8    6.225  5.75   5.225  6.3    6.55   5.7    5.85   6.55\n",
1103         "  5.775  5.825  6.175  6.1    5.8    6.425  6.05   6.025  6.175  6.55\n",
1104         "  6.175  6.35   6.725  6.125  7.075  5.725  5.925  6.15   6.075  5.75\n",
1105         "  5.975  5.725  6.3    5.9    6.75   5.925  7.225  6.15   5.95   6.275  5.7\n",
1106         "  6.1    6.825  5.975  6.725  5.7    6.25   6.4    7.05   5.9  ]\n"
1107        ]
1108       }
1109      ],
1110      "prompt_number": 25
1111     },
1112     {
1113      "cell_type": "markdown",
1114      "metadata": {
1115       "cell_tags": []
1116      },
1117      "source": [
1118       "which is the average inflammation per patient across all days."
1119      ]
1120     },
1121     {
1122      "cell_type": "markdown",
1123      "metadata": {
1124       "cell_tags": [
1125        "challenges"
1126       ]
1127      },
1128      "source": [
1129       "#### Challenges\n",
1130       "\n",
1131       "A subsection of an array is called a [slice](../../gloss.html#slice).\n",
1132       "We can take slices of character strings as well:"
1133      ]
1134     },
1135     {
1136      "cell_type": "code",
1137      "collapsed": false,
1138      "input": [
1139       "element = 'oxygen'\n",
1140       "print 'first three characters:', element[0:3]\n",
1141       "print 'last three characters:', element[3:6]"
1142      ],
1143      "language": "python",
1144      "metadata": {
1145       "cell_tags": [
1146        "challenges"
1147       ]
1148      },
1149      "outputs": [
1150       {
1151        "output_type": "stream",
1152        "stream": "stdout",
1153        "text": [
1154         "first three characters: oxy\n",
1155         "last three characters: gen\n"
1156        ]
1157       }
1158      ],
1159      "prompt_number": 26
1160     },
1161     {
1162      "cell_type": "markdown",
1163      "metadata": {
1164       "cell_tags": [
1165        "challenges"
1166       ]
1167      },
1168      "source": [
1169       "1.  What is the value of `element[:4]`?\n",
1170       "    What about `element[4:]`?\n",
1171       "    Or `element[:]`?\n",
1172       "\n",
1173       "1.  What is `element[-1]`?\n",
1174       "    What is `element[-2]`?\n",
1175       "    Given those answers,\n",
1176       "    explain what `element[1:-1]` does.\n",
1177       "\n",
1178       "1.  The expression `element[3:3]` produces an [empty string](../../gloss.html#empty-string),\n",
1179       "    i.e., a string that contains no characters.\n",
1180       "    If `data` holds our array of patient data,\n",
1181       "    what does `data[3:3, 4:4]` produce?\n",
1182       "    What about `data[3:3, :]`?"
1183      ]
1184     },
1185     {
1186      "cell_type": "heading",
1187      "level": 3,
1188      "metadata": {
1189       "cell_tags": []
1190      },
1191      "source": [
1192       "Plotting"
1193      ]
1194     },
1195     {
1196      "cell_type": "markdown",
1197      "metadata": {
1198       "cell_tags": []
1199      },
1200      "source": [
1201       "The mathematician Richard Hamming once said,\n",
1202       "\"The purpose of computing is insight, not numbers,\"\n",
1203       "and the best way to develop insight is often to visualize data.\n",
1204       "Visualization deserves an entire lecture (or course) of its own,\n",
1205       "but we can explore a few features of Python's `matplotlib` here.\n",
1206       "First,\n",
1207       "let's tell the IPython Notebook that we want our plots displayed inline,\n",
1208       "rather than in a separate viewing window:"
1209      ]
1210     },
1211     {
1212      "cell_type": "code",
1213      "collapsed": false,
1214      "input": [
1215       "%matplotlib inline"
1216      ],
1217      "language": "python",
1218      "metadata": {
1219       "cell_tags": []
1220      },
1221      "outputs": [],
1222      "prompt_number": 27
1223     },
1224     {
1225      "cell_type": "markdown",
1226      "metadata": {
1227       "cell_tags": []
1228      },
1229      "source": [
1230       "The `%` at the start of the line signals that this is a command for the notebook,\n",
1231       "rather than a statement in Python.\n",
1232       "Next,\n",
1233       "we will import the `pyplot` module from `matplotlib`\n",
1234       "and use two of its functions to create and display a heat map of our data:"
1235      ]
1236     },
1237     {
1238      "cell_type": "code",
1239      "collapsed": false,
1240      "input": [
1241       "from matplotlib import pyplot\n",
1242       "pyplot.imshow(data)\n",
1243       "pyplot.show()"
1244      ],
1245      "language": "python",
1246      "metadata": {
1247       "cell_tags": []
1248      },
1249      "outputs": [
1250       {
1251        "metadata": {},
1252        "output_type": "display_data",
1253        "png": "iVBORw0KGgoAAAANSUhEUgAAALIAAAD+CAYAAACeEF9/AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
1254        "text": [
1255         "<matplotlib.figure.Figure at 0x10600f5d0>"
1256        ]
1257       }
1258      ],
1259      "prompt_number": 28
1260     },
1261     {
1262      "cell_type": "markdown",
1263      "metadata": {
1264       "cell_tags": []
1265      },
1266      "source": [
1267       "Blue regions in this heat map are low values, while red shows high values.\n",
1268       "As we can see,\n",
1269       "inflammation rises and falls over a 40-day period.\n",
1270       "Let's take a look at the average inflammation over time:"
1271      ]
1272     },
1273     {
1274      "cell_type": "code",
1275      "collapsed": false,
1276      "input": [
1277       "ave_inflammation = data.mean(axis=0)\n",
1278       "pyplot.plot(ave_inflammation)\n",
1279       "pyplot.show()"
1280      ],
1281      "language": "python",
1282      "metadata": {
1283       "cell_tags": []
1284      },
1285      "outputs": [
1286       {
1287        "metadata": {},
1288        "output_type": "display_data",
1289        "png": "iVBORw0KGgoAAAANSUhEUgAAAXEAAAEACAYAAABF+UbAAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n9v8BdEzZ6MQ88gYAAAAASUVORK5CYII=\n",
1290        "text": [
1291         "<matplotlib.figure.Figure at 0x106026750>"
1292        ]
1293       }
1294      ],
1295      "prompt_number": 29
1296     },
1297     {
1298      "cell_type": "markdown",
1299      "metadata": {
1300       "cell_tags": []
1301      },
1302      "source": [
1303       "Here,\n",
1304       "we have put the average per day across all patients in the variable `ave_inflammation`,\n",
1305       "then asked `pyplot` to create and display a line graph of those values.\n",
1306       "The result is roughly a linear rise and fall,\n",
1307       "which is suspicious:\n",
1308       "based on other studies,\n",
1309       "we expect a sharper rise and slower fall.\n",
1310       "Let's have a look at two other statistics:"
1311      ]
1312     },
1313     {
1314      "cell_type": "code",
1315      "collapsed": false,
1316      "input": [
1317       "print 'maximum inflammation per day'\n",
1318       "pyplot.plot(data.max(axis=0))\n",
1319       "pyplot.show()\n",
1320       "\n",
1321       "print 'minimum inflammation per day'\n",
1322       "pyplot.plot(data.min(axis=0))\n",
1323       "pyplot.show()"
1324      ],
1325      "language": "python",
1326      "metadata": {
1327       "cell_tags": []
1328      },
1329      "outputs": [
1330       {
1331        "output_type": "stream",
1332        "stream": "stdout",
1333        "text": [
1334         "maximum inflammation per day\n"
1335        ]
1336       },
1337       {
1338        "metadata": {},
1339        "output_type": "display_data",
1340        "png": "iVBORw0KGgoAAAANSUhEUgAAAXIAAAEACAYAAACuzv3DAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAABJRU5ErkJggg==\n",
1341        "text": [
1342         "<matplotlib.figure.Figure at 0x106036bd0>"
1343        ]
1344       },
1345       {
1346        "output_type": "stream",
1347        "stream": "stdout",
1348        "text": [
1349         "minimum inflammation per day\n"
1350        ]
1351       },
1352       {
1353        "metadata": {},
1354        "output_type": "display_data",
1355        "png": "iVBORw0KGgoAAAANSUhEUgAAAWwAAAEACAYAAACXqUyYAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nTkSuQmCC\n",
1356        "text": [
1357         "<matplotlib.figure.Figure at 0x106033550>"
1358        ]
1359       }
1360      ],
1361      "prompt_number": 30
1362     },
1363     {
1364      "cell_type": "markdown",
1365      "metadata": {
1366       "cell_tags": []
1367      },
1368      "source": [
1369       "The maximum value rises and falls perfectly smoothly,\n",
1370       "while the minimum seems to be a step function.\n",
1371       "Neither result seems particularly likely,\n",
1372       "so either there's a mistake in our calculations\n",
1373       "or something is wrong with our data."
1374      ]
1375     },
1376     {
1377      "cell_type": "markdown",
1378      "metadata": {
1379       "cell_tags": [
1380        "challenges"
1381       ]
1382      },
1383      "source": [
1384       "#### Challenges\n",
1385       "\n",
1386       "1.  Why do all of our plots stop just short of the upper end of our graph?\n",
1387       "    Why are the vertical lines in our plot of the minimum inflammation per day not vertical?\n",
1388       "\n",
1389       "1.  Create a plot showing the standard deviation of the inflammation data for each day across all patients."
1390      ]
1391     },
1392     {
1393      "cell_type": "heading",
1394      "level": 3,
1395      "metadata": {
1396       "cell_tags": []
1397      },
1398      "source": [
1399       "Wrapping Up"
1400      ]
1401     },
1402     {
1403      "cell_type": "markdown",
1404      "metadata": {
1405       "cell_tags": []
1406      },
1407      "source": [
1408       "It's very common to create an [alias](../../gloss.html#alias-library) for a library when importing it\n",
1409       "in order to reduce the amount of typing we have to do.\n",
1410       "Here are our three plots side by side using aliases for `numpy` and `pyplot`:"
1411      ]
1412     },
1413     {
1414      "cell_type": "code",
1415      "collapsed": false,
1416      "input": [
1417       "import numpy as np\n",
1418       "from matplotlib import pyplot as plt\n",
1419       "\n",
1420       "data = np.loadtxt(fname='inflammation-01.csv', delimiter=',')\n",
1421       "\n",
1422       "plt.figure(figsize=(10.0, 3.0))\n",
1423       "\n",
1424       "plt.subplot(1, 3, 1)\n",
1425       "plt.ylabel('average')\n",
1426       "plt.plot(data.mean(0))\n",
1427       "\n",
1428       "plt.subplot(1, 3, 2)\n",
1429       "plt.ylabel('max')\n",
1430       "plt.plot(data.max(0))\n",
1431       "\n",
1432       "plt.subplot(1, 3, 3)\n",
1433       "plt.ylabel('min')\n",
1434       "plt.plot(data.min(0))\n",
1435       "\n",
1436       "plt.tight_layout()\n",
1437       "plt.show()"
1438      ],
1439      "language": "python",
1440      "metadata": {
1441       "cell_tags": []
1442      },
1443      "outputs": [
1444       {
1445        "metadata": {},
1446        "output_type": "display_data",
1447        "png": "iVBORw0KGgoAAAANSUhEUgAAAskAAADSCAYAAAC4u12cAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nD8Se36QdbJ39AAAAAElFTkSuQmCC\n",
1448        "text": [
1449         "<matplotlib.figure.Figure at 0x106062fd0>"
1450        ]
1451       }
1452      ],
1453      "prompt_number": 31
1454     },
1455     {
1456      "cell_type": "markdown",
1457      "metadata": {
1458       "cell_tags": []
1459      },
1460      "source": [
1461       "The first two lines re-load our libraries as `np` and `plt`,\n",
1462       "which are the aliases most Python programmers use.\n",
1463       "The call to `loadtxt` reads our data,\n",
1464       "and the rest of the program tells the plotting library\n",
1465       "how large we want the figure to be,\n",
1466       "that we're creating three sub-plots,\n",
1467       "what to draw for each one,\n",
1468       "and that we want a tight layout.\n",
1469       "(Perversely,\n",
1470       "if we leave out that call to `plt.tight_layout()`,\n",
1471       "the graphs will actually be squeezed together more closely.)"
1472      ]
1473     },
1474     {
1475      "cell_type": "markdown",
1476      "metadata": {
1477       "cell_tags": [
1478        "challenges"
1479       ]
1480      },
1481      "source": [
1482       "#### Challenges\n",
1483       "\n",
1484       "1.  Modify the program to display the three plots on top of one another instead of side by side."
1485      ]
1486     },
1487     {
1488      "cell_type": "markdown",
1489      "metadata": {
1490       "cell_tags": [
1491        "keypoints"
1492       ]
1493      },
1494      "source": [
1495       "#### Key Points\n",
1496       "\n",
1497       "*   Import a library into a program using `import libraryname`.\n",
1498       "*   Use the `numpy` library to work with arrays in Python.\n",
1499       "*   Use `variable = value` to assign a value to a variable in order to record it in memory.\n",
1500       "*   Variables are created on demand whenever a value is assigned to them.\n",
1501       "*   Use `print something` to display the value of `something`.\n",
1502       "*   The expression `array.shape` gives the shape of an array.\n",
1503       "*   Use `array[x, y]` to select a single element from an array.\n",
1504       "*   Array indices start at 0, not 1.\n",
1505       "*   Use `low:high` to specify a slice that includes the indices from `low` to `high-1`.\n",
1506       "*   All the indexing and slicing that works on arrays also works on strings.\n",
1507       "*   Use `# some kind of explanation` to add comments to programs.\n",
1508       "*   Use `array.mean()`, `array.max()`, and `array.min()` to calculate simple statistics.\n",
1509       "*   Use `array.mean(axis=0)` or `array.mean(axis=1)` to calculate statistics across the specified axis.\n",
1510       "*   Use the `pyplot` library from `matplotlib` for creating simple visualizations."
1511      ]
1512     },
1513     {
1514      "cell_type": "markdown",
1515      "metadata": {
1516       "cell_tags": []
1517      },
1518      "source": [
1519       "#### Next Steps\n",
1520       "\n",
1521       "Our work so far has convinced us that something's wrong with our first data file.\n",
1522       "We would like to check the other 11 the same way,\n",
1523       "but typing in the same commands repeatedly is tedious and error-prone.\n",
1524       "Since computers don't get bored (that we know of),\n",
1525       "we should create a way to do a complete analysis with a single command,\n",
1526       "and then figure out how to repeat that step once for each file.\n",
1527       "These operations are the subjects of the next two lessons."
1528      ]
1529     }
1530    ],
1531    "metadata": {}
1532   }
1533  ]
1534 }