1
00:00:00,910 --> 00:00:02,750
- To finish up our discussion of

2
00:00:02,750 --> 00:00:04,960
the Titanic data set for now,

3
00:00:04,960 --> 00:00:08,820
we'd like to take a look
at how we can visualize

4
00:00:08,820 --> 00:00:10,540
a little bit of this data,

5
00:00:10,540 --> 00:00:11,670
and for that purpose,

6
00:00:11,670 --> 00:00:14,880
we're going to take
advantage of a built in

7
00:00:14,880 --> 00:00:18,520
Pandas capability for
displaying a histogram

8
00:00:18,520 --> 00:00:23,010
based on specifically the
age column in this case.

9
00:00:23,010 --> 00:00:25,210
So in order to do that,

10
00:00:25,210 --> 00:00:28,490
we need to enable Matplotlib support.

11
00:00:28,490 --> 00:00:31,990
Pandas uses Matplotlib to
implement its graphics.

12
00:00:31,990 --> 00:00:36,590
We're in a curin IPython
session in which I did not

13
00:00:36,590 --> 00:00:40,430
enable Matplotlib support
as I launched the session,

14
00:00:40,430 --> 00:00:43,360
so here we want to talk about using

15
00:00:43,360 --> 00:00:46,373
the magic called Matplotlib.

16
00:00:47,590 --> 00:00:49,700
And what Matplotlib,

17
00:00:49,700 --> 00:00:52,050
I'm just making sure I
spelled it correctly here.

18
00:00:52,050 --> 00:00:55,690
And that's going to
enable Matplotlib support

19
00:00:55,690 --> 00:00:57,210
in the current session,

20
00:00:57,210 --> 00:01:01,760
so even if you didn't launch
Ipython with Matplotlib support

21
00:01:01,760 --> 00:01:05,670
you can enable it after the
fact by using this magic.

22
00:01:05,670 --> 00:01:08,920
So now that we have that set up,

23
00:01:08,920 --> 00:01:12,330
we can go ahead and create a histogram,

24
00:01:12,330 --> 00:01:15,770
and to do that, we'll
use a very simple method

25
00:01:15,770 --> 00:01:18,790
call hist stands for histogram,

26
00:01:18,790 --> 00:01:22,400
and when you invoke that on a data frame,

27
00:01:22,400 --> 00:01:24,890
it's going to create histograms for

28
00:01:24,890 --> 00:01:28,540
all of the numeric columns by default.

29
00:01:28,540 --> 00:01:29,900
So there's only one,

30
00:01:29,900 --> 00:01:32,410
which means it's only going
to create one histogram,

31
00:01:32,410 --> 00:01:34,950
and in this case we'll
hit enter and let that

32
00:01:34,950 --> 00:01:35,920
create the histogram

33
00:01:35,920 --> 00:01:38,290
which will pop open in
a separate window here.

34
00:01:38,290 --> 00:01:41,210
And now what you see is a histogram

35
00:01:41,210 --> 00:01:45,310
of everything created for you
auto-magically, if you will.

36
00:01:45,310 --> 00:01:46,770
On the left hand column,

37
00:01:46,770 --> 00:01:51,020
we have the number of
passengers in a given age range,

38
00:01:51,020 --> 00:01:53,890
and down along the bottom
we have the age ranges.

39
00:01:53,890 --> 00:01:56,980
So you can see this first block here

40
00:01:56,980 --> 00:02:01,570
represents what looks like
about zero through eight or so,

41
00:02:01,570 --> 00:02:05,410
so there were somewhere
around maybe 70 people

42
00:02:05,410 --> 00:02:08,320
who were in the zero to eight range.

43
00:02:08,320 --> 00:02:09,700
In this next block here,

44
00:02:09,700 --> 00:02:11,770
maybe we had about 60 people

45
00:02:11,770 --> 00:02:16,370
who were in the eight to
approximately 16 or 17 range.

46
00:02:16,370 --> 00:02:18,300
In this next block here,

47
00:02:18,300 --> 00:02:23,280
we had over 250 people, maybe 275 or so,

48
00:02:23,280 --> 00:02:26,390
280 people that were in approximately

49
00:02:26,390 --> 00:02:31,010
the 17 to maybe 24 range, et cetera,

50
00:02:31,010 --> 00:02:34,840
so you can see we get this
kind of contiguous presentation

51
00:02:34,840 --> 00:02:37,348
of ranges of values and

52
00:02:37,348 --> 00:02:41,230
the number of people inside those ranges.

53
00:02:41,230 --> 00:02:43,730
So, again, this is a key aspect of

54
00:02:43,730 --> 00:02:45,660
getting to know your data,

55
00:02:45,660 --> 00:02:49,070
to visualize that data,

56
00:02:49,070 --> 00:02:51,900
and sometimes by looking
at it in a picture,

57
00:02:51,900 --> 00:02:56,580
you can understand the data
a little bit better as well.

58
00:02:56,580 --> 00:03:00,010
So we will of course do
lots of visualizations

59
00:03:00,010 --> 00:03:03,010
in subsequent examples and
we just wanted to give you

60
00:03:03,010 --> 00:03:05,120
a basic demonstration here,

61
00:03:05,120 --> 00:03:09,480
and of course, Pandas has other
visualization capabilities

62
00:03:09,480 --> 00:03:11,930
above and beyond simple histograms

63
00:03:11,930 --> 00:03:13,863
that we're showing in this example.