1 .. Author              : Nishanth

     2    Internal Reviewer 1 : Anoop

     3    Internal Reviewer 2 : Madhu

     4    External Reviewer   :

     5 

     6 .. Prerequisites: using ``plot`` command

     7 

     8 Hello friends and welcome to the tutorial on Embellishing Plots.

     9 

    10 {{{ Show the slide containing title }}}

    11 

    12 {{{ Show the slide containing the outline }}}

    13 

    14 In this tutorial, we shall look at how to modify the colour, thickness and 

    15 linestyle of the plot. We shall then learn how to add title to the plot and 

    16 then look at adding labels to x and y axes. we shall also look at adding 

    17 annotations to the plot and setting the limits of axes.

    18 

    19 Let us start ipython with pylab loaded, by typing on the terminal

    20 

    21 {{{ shift to terminal and type ipython -pylab }}}

    22 

    23 ::

    24 

    25     ipython -pylab

    26 

    27 .. #[madhu: I feel the instructions should precede the actual action,

    28 

    29 since while recording we need to know before hand what we need to do]

    30 

    31 We shall first make a simple plot and start decorating it.

    32 

    33 .. #[madhu: start decorating it should be fine, with is not necessary]

    34 

    35 ::

    36 

    37     x = linspace(-2, 4, 20)

    38     plot(x, sin(x))

    39 

    40 .. #[madhu: Standard is to choose between -50 to 50 or 0 to 50 with 100

    41      points right?]

    42 

    43 As we can see, the default colour and the default thickness of the

    44 line is as decided by pylab. Wouldn't be nice if we could control

    45 these parameters in the plot? This is possible by passing additional

    46 arguments to the plot command.

    47 

    48 .. #[[Anoop: I think it will be good to rephrase the sentence]]

    49 .. #[madhu: Why "you" here? Shouldn't this be "we" as decided? Also I

    50      added "the default" check the diff]

    51 

    52 The additional argument that we shall be passing in here now is the

    53 colour argument. We shall first clear the figure and plot the same in

    54 red colour. Hence

    55 

    56 .. #[Madhu: Note the diff for changes]

    57  ::

    58 

    59     clf()

    60     plot(x, sin(x), 'r')

    61 

    62 As we can see we have the same plot but now in red colour.

    63 

    64 .. #[Madhu: diff again]

    65 

    66 To alter the thickness of the line, we use the ``linewidth`` argument in the plot

    67 command. Hence

    68 ::

    69 

    70     plot(x, cos(x), linewidth=2)

    71 

    72 produces a plot with a thicker line, to be more precise plot with line

    73 thickness 2.

    74 

    75 .. #[[Anoop: I guess it will be good if you say that it affects the

    76    same plot, as you have not cleared the figure]]

    77 .. #[Madhu: To Anoop, not necessary I feel since they can see it?]

    78 

    79 {{{ Show the plot and compare the sine and cos plots }}}

    80 

    81 {{{ Pause here and try out the following exercises }}}

    82 

    83 .. #[[Anoop: is the above a context switch for the person who does the

    84    recording, other wise if it an instruction to the person viewing

    85    the video, then I guess the three braces can be removed.]]

    86 

    87 %% 1 %% Plot sin(x) in blue colour and with linewidth as 3

    88 

    89 {{{ continue from paused state }}}

    90 

    91 A combination of colour and linewidth would do the job for us. Hence

    92 ::

    93 

    94     clf()

    95     plot(x, sin(x), 'b', linewidth=3)

    96 

    97 .. #[[Anoop: add clf()]]

    98 

    99 produces the required plot

   100 

   101 .. #[Nishanth]: I could not think of a SIMPLE recipe approach for

   102              introducing linestyle. Hence the naive approach.

   103 

   104 .. #[[Anoop: I guess the recipe is fine, but would be better if you

   105    add the problem statement rather than just saying "let's do a simple

   106    plot"]]

   107 

   108 .. #[Madhu: It is good enough.]

   109 

   110 Occasionally we would also want to alter the style of line. Sometimes

   111 all we want is just a bunch of points not joined. This is possible by

   112 passing the linestyle argument along with or instead of the colour

   113 argument. Hence ::

   114 

   115     clf()

   116     plot(x, sin(x), '.')

   117 

   118 produces a plot with only points.

   119 

   120 To produce the same plot but now in blue colour, we do

   121 ::

   122 

   123     clf()

   124     plot(x, sin(x), 'b.')

   125 

   126 Other available options can be seen in the documentation of plot.

   127 ::

   128 

   129     plot?

   130 

   131 {{{ Run through the documentation and show the options available }}}

   132 

   133 {{{ Show the options available for line style and colors }}}

   134 

   135 .. #[Madhu: The script needs to tell what needs to be shown or

   136      explained.]

   137 

   138 {{{ Pause here and try out the following exercises }}}

   139 

   140 .. #[[Anoop: same question as above, should it be read out?]]

   141 

   142 %% 2 %% Plot the sine curve with green filled circles.

   143 

   144 {{{ continue from paused state }}}

   145 

   146 All we have to do is use a combination of linestyle and colour to acheive this.

   147 Hence

   148 ::

   149 

   150     clf()

   151     plot(x, cos(x), 'go')

   152 

   153 produces the required plot.

   154 

   155 {{{ Pause here and try out the following exercises }}}

   156 

   157 %% 3 %% Plot the curve of x vs tan(x) in red dashed line and linewidth 3

   158 

   159 {{{ continue from paused state }}}

   160 

   161 .. #[Madhu: I did not understand the question]

   162 

   163 Now that we know how to produce a bare minimum plot with colour, style

   164 and thickness of our interest, we shall look at decorating the plot.

   165 

   166 Let us start with a plot of the function -x^2 + 4x - 5.

   167 ::

   168 

   169     plot(x, -x*x + 4*x - 5, 'r', linewidth=2)

   170 

   171 {{{ Show the plot window and switch back to terminal }}}

   172 

   173 We now have the plot in a colour and linewidth of our interest. As you can see,

   174 the figure does not have any description describing the plot.

   175 

   176 .. #[Madhu: Added "not". See the diff]

   177 

   178 We will now add a title to the plot by using the ``title`` command.

   179 ::

   180 

   181     title("Parabolic function -x^2+4x-5") 

   182 

   183 {{{ Show the plot window and point to the title }}}

   184 

   185 The figure now has a title which describes what the plot is. The

   186 ``title`` command as you can see, takes a string as an argument and sets

   187 the title accordingly.

   188 

   189 .. #[Madhu: See the diff]

   190 

   191 The formatting in title is messed and it does not look clean. You can imagine

   192 what would be the situation if there were fractions and more complex functions

   193 like log and exp. Wouldn't it be good if there was LaTex like formatting?

   194 

   195 That is also possible by adding a $ sign before and after the part of the 

   196 string that should be in LaTex style.

   197 

   198 for instance, we can use

   199 ::

   200 

   201     title("Parabolic function $-x^2+4x-5$")

   202 

   203 and we get the polynomial formatted properly.

   204 

   205 .. #[Nishanth]: Unsure if I have to give this exercise since enclosing the whole

   206              string in LaTex style is not good

   207 

   208 .. #[[Anoop: I guess you can go ahead with the LaTex thing, it's

   209      cool!]]

   210 .. #[Madhu: Instead of saying LaTeX style you can say Typeset math

   211      since that is how it is called as. I am not sure as well. It

   212      doesn't really solve the purpose]

   213 

   214 {{{ Pause here and try out the following exercises }}}

   215 

   216 %% 4 %% Change the title of the figure such that the whole title is formatted

   217         in LaTex style

   218 

   219 {{{ continue from the paused state }}}

   220 

   221 The solution is to enclose the whole string in between $. Hence,

   222 ::

   223 

   224     title("$Parabolic function -x^2+4x-5$")

   225 

   226 gives a title that looks neatly formatted.

   227 

   228 Although we have title, the plot is not complete without labelling x

   229 and y axes. Hence we shall label x-axis to "x" and y-axis to "f(x)" ::

   230 

   231     xlabel("x")

   232 

   233 {{{ Switch to plot window and show the xlabel }}}

   234 

   235 As you can see, ``xlabel`` command takes a string as an argument,

   236 similar to the ``title`` command and sets it as the label to x-axis.

   237 

   238 .. #[See the diff]

   239 

   240 Similarly,

   241 ::

   242 

   243     ylabel("f(x)")

   244 

   245 sets the name of the y-axis as "f(x)"

   246 

   247 {{{ Show the plot window and point to ylabel and switch back to the terminal }}}

   248 

   249 {{{ Pause here and try out the following exercises }}}

   250 

   251 %% 5 %% Set the x and y labels as "x" and "f(x)" in LaTex style.

   252 

   253 {{{ continue from paused state }}}

   254 

   255 Since we need LaTex style formatting, all we have to do is enclose the string

   256 in between two $. Hence,

   257 ::

   258 

   259     xlabel("$x$")

   260     yalbel("$f(x)$")

   261 

   262 does the job for us.

   263 

   264 {{{ Show the plot window with clean labels }}}

   265 

   266 The plot is now almost complete. Except that we have still not seen how to 

   267 name the points. For example the point (2, -1) is the local maxima. We would

   268 like to name the point accordingly. We can do this by using

   269 ::

   270 

   271     annotate("local maxima", xy=(2, -1))

   272 

   273 {{{ Show the annotation that has appeared on the plot }}}

   274 

   275 As you can see, the first argument to ``annotate`` command is the name we would

   276 like to mark the point as and the second argument is the co-ordinates of the

   277 point at which the name should appear. It is a sequence containing two numbers.

   278 The first is x co-ordinate and second is y co-ordinate.

   279 

   280 .. #[[Anoop: I think we should tell explicitely that xy takes a

   281    sequence or a tuple]]

   282 .. #[Madhu: Agreed to what anoop says and also that xy= is the point

   283      part should be rephrased I think.]

   284 

   285 {{{ Pause here and try out the following exercises }}}

   286 

   287 %% 6 %% Make an annotation called "root" at the point (-4, 0)

   288         What happens to the first annotation ?

   289 

   290 {{{ continue from paused state }}}

   291 

   292 As we can see, every annotate command makes a new annotation on the figure.

   293 

   294 Now we have everything we need to decorate a plot. but the plot would be

   295 incomplete if we can not set the limits of axes. This is possible using the

   296 button on the plot window.

   297 

   298 we shall look at how to get and set them from the script.

   299 ::

   300 

   301     xlim()

   302     ylim()

   303 

   304 We see that ``xlim`` function returns the current x axis limits and ylim

   305 function returns the current y-axis limits.

   306 

   307 Let us look at how to set the limits.

   308 ::

   309 

   310     xlim(-4, 5)

   311 

   312 We see the limits of x-axis are now set to -4 and 5.

   313 Similarly

   314 ::

   315 

   316     ylim(-15, 2)

   317 

   318 sets the limits of y-axis appropriately.

   319 

   320 {{{ Pause here and try out the following exercises }}}

   321 

   322 %% 7 %% Set the limits of axes such that the area of interest is the rectangle

   323         (-1, -15) and (3, 0)

   324 

   325 {{{ continue from paused state }}}

   326 

   327 As we can see, the lower upper limits of x-axis in the question are -1 and 3.

   328 The limits of y-axis are -15 and 0.

   329 

   330 ::

   331 

   332     xlim(-1, 3)

   333     ylim(-15, 0)

   334 

   335 Gives us the required rectangle.

   336 

   337 {{{ Show summary slide }}}

   338 

   339 we have looked at 

   340 

   341  * Modifying the attributes of plot by passing additional arguments

   342  * How to add title

   343  * How to incorporate LaTex style formatting

   344  * How to label x and y axes

   345  * How to add annotations

   346  * How to set the limits of axes

   347 

   348 {{{ Show the "sponsored by FOSSEE" slide }}}

   349 

   350 .. #[Nishanth]: Will add this line after all of us fix on one.

   351 

   352 This tutorial was created as a part of FOSSEE project, NME ICT, MHRD India

   353 

   354 Hope you have enjoyed and found it useful.

   355 Thankyou

   356 

   357 Questions

   358 =========

   359 

   360  1. Draw a plot of cosine graph between -2pi to 2pi with line thickness 4

   361 

   362     Answer::

   363 

   364     x = linspace(-2*pi, 2*pi)

   365     plot(x, cos(x), linewidth=4)

   366 

   367  2. Draw a plot of the polynomial x^2-5x+6 in the range 0 to 5 in blue dotted

   368     line

   369 

   370     Answer::

   371 

   372     x = linspace(-2*pi, 2*pi)

   373     plot(x, x**2 - 5*x + 6, 'r.')

   374 

   375  3. Which marker is used to get circles

   376 

   377     a. '.'

   378     #. '^'

   379     #. 'o'

   380     #. '--'

   381 

   382  4. What does the '^' marker produce

   383 

   384     Answer: Triangle up marker

   385 

   386  5. How do you set the title as x^2-5x+6 in LaTex style formatting

   387 

   388     Answer: title("$x^2-5x+6$")

   389 

   390 6. What happens when the following code is executed::

   391 

   392     xlabel("First label")

   393     xlabel("Second label")

   394 

   395    Answer: The label of x-axis is set to "Second label"

   396 

   397  7. Read thorugh the documentation and find out is there a way to modify the

   398     alignment of text in the command ``ylabel``

   399 

   400    a. Yes

   401    #. No

   402 

   403    Answer: No

   404 

   405  8. How to add the annotation "Maxima" at the point (1, 2)

   406 

   407    Answer: annotate("Maxima", xy=(1, 2))

   408 

   409  9. Is the command ``annotate("max", (1, 2))`` same as ``annotate("max",

   410     xy=(1, 2)``

   411 

   412     a. True

   413     b. False

   414 

   415     Answer: True

   416 

   417  10. When a new annotation is made at a point, what happens to the old one

   418 

   419     a. It is replaced

   420     b. It is overwritten

   421     c. The new annotation is combined with old one

   422 

   423     Answer: It is overwritten

   424 

   425  11. What happens when xlim is used without arguments

   426 

   427     Answer: It gives the current limits of x-axis

   428 

   429  12. What happens when ``ylim(0, 5)`` is used

   430 

   431     Answer: It sets the lower and upper limits of y-axis to 0 and 5

   432 

   433  13. Draw a cosine plot from 0 to 2*pi with green dots. annotate the origin as

   434      "origin" and set x and y labels to "x" and cos(x) and x limits to 0 and

   435      2pi and y limits to -1.2 and 1.2

   436 

   437     Answer::

   438 

   439       x = linspace(0, 2*pi)

   440       plot(x, cos(x), 'g.')

   441       annotate("origin", (0, 0))

   442       xlabel("$x$")

   443       ylabel("$cos(x)$")

   444       xlim(0, 2*pi)

   445       ylim(-1.2, 1.2)

   446