1
00:00:00,266 --> 00:00:02,233
And now we're going to tackle very quickly

2
00:00:02,233 --> 00:00:05,700
the very first step of this toolkit,
which is the libraries.

3
00:00:05,933 --> 00:00:09,266
How to import the libraries,
how to have them ready

4
00:00:09,300 --> 00:00:12,433
any time we start building a new machine
learning model will.

5
00:00:12,433 --> 00:00:13,133
There we go.

6
00:00:13,133 --> 00:00:14,500
I will show you this right away.

7
00:00:14,500 --> 00:00:16,166
It's actually very simple.

8
00:00:16,166 --> 00:00:22,000
We're going to import three libraries now
which are numpy, matplotlib, and pandas.

9
00:00:22,366 --> 00:00:25,933
NumPy will allow us to work with arrays
because indeed you will see

10
00:00:25,933 --> 00:00:30,300
that your future machine learning models
will expect some arrays as inputs.

11
00:00:30,466 --> 00:00:33,600
And therefore we need a library
to work with these arrays.

12
00:00:33,600 --> 00:00:34,800
And that's numpy.

13
00:00:34,800 --> 00:00:38,633
Then we'll import matplotlib,
which is the library that will allow us

14
00:00:38,633 --> 00:00:40,266
to plot some very nice charts.

15
00:00:40,266 --> 00:00:43,800
You will see that we will actually
many charts and graphs in this course.

16
00:00:44,133 --> 00:00:48,866
And finally pandas, which will allow us
to not only import the data set

17
00:00:48,866 --> 00:00:52,333
but also create the matrix of features
and the dependent variable vector.

18
00:00:52,333 --> 00:00:54,900
I will explain, of course,
all these concepts later.

19
00:00:54,900 --> 00:00:55,233
All right.

20
00:00:55,233 --> 00:00:59,366
So let's import these libraries
in order to import a library in Python.

21
00:00:59,366 --> 00:01:02,466
It's very simple. You start with import.

22
00:01:03,000 --> 00:01:06,300
That's just a command
that will allow to import a library

23
00:01:06,300 --> 00:01:09,966
or even a function or any type of modules.

24
00:01:10,100 --> 00:01:13,833
By the way, a library is
an ensemble of modules containing

25
00:01:13,833 --> 00:01:18,000
functions and classes with which you can
perform some actions and operations.

26
00:01:18,166 --> 00:01:22,600
For example, the most well-known library
in data science is scikit learn,

27
00:01:22,600 --> 00:01:25,933
and this is a library that contains
actually all the machine learning models

28
00:01:25,933 --> 00:01:30,133
that you can build by simply
creating an object of some classes.

29
00:01:30,133 --> 00:01:33,900
So don't worry about all this
now we will see that in great detail

30
00:01:33,900 --> 00:01:35,866
each time we build a machine
learning models.

31
00:01:35,866 --> 00:01:36,866
But there you go.

32
00:01:36,866 --> 00:01:39,600
You will see that
libraries will be super useful for us

33
00:01:39,600 --> 00:01:42,000
in order to build our machine
learning models.

34
00:01:42,000 --> 00:01:43,566
So there we go. Import.

35
00:01:43,566 --> 00:01:47,000
Then you need to follow by the name
of the library you want to import.

36
00:01:47,200 --> 00:01:50,400
And the first one we want to import
is called numpy.

37
00:01:50,700 --> 00:01:53,733
And then what we usually do
is we add a shortcut,

38
00:01:53,733 --> 00:01:58,133
because each time we will use
a function of the numpy library.

39
00:01:58,300 --> 00:02:00,933
Well we'll have to call first numpy.

40
00:02:00,933 --> 00:02:02,700
And in order to call it faster.

41
00:02:02,700 --> 00:02:06,200
Well we can add a shortcut here
so that each time we want to call number

42
00:02:06,400 --> 00:02:08,833
we will actually call it
with its shortcut.

43
00:02:08,833 --> 00:02:12,266
And the shortcut
we usually use for numpy is NP.

44
00:02:12,266 --> 00:02:15,300
And as you can see, to add the shortcut
you simply need to add as

45
00:02:15,300 --> 00:02:17,166
and then the shortcut name. All right.

46
00:02:17,166 --> 00:02:20,233
So each time we will call numpy
we will actually call NP.

47
00:02:20,966 --> 00:02:23,466
Then next library import.

48
00:02:23,466 --> 00:02:25,800
So that next library is matplotlib.

49
00:02:25,800 --> 00:02:29,200
And actually as I told you
a library is an ensemble of modules.

50
00:02:29,200 --> 00:02:32,500
And here we're interested
in a particular module called Pyplot.

51
00:02:32,500 --> 00:02:35,633
And that's the module
that allows us to plot very nice charts.

52
00:02:35,966 --> 00:02:39,733
And so here
we're not only going to import matplotlib,

53
00:02:39,866 --> 00:02:44,333
but particularly, well the pyplot module.

54
00:02:44,333 --> 00:02:46,533
And as you can see
in order to get this Pyplot module,

55
00:02:46,533 --> 00:02:50,400
I added a dot here which allows you
to access the different modules

56
00:02:50,400 --> 00:02:54,166
of the matplotlib library
and the module we chose is Pyplot.

57
00:02:54,166 --> 00:02:57,033
And again
we're going to add a little shortcut here,

58
00:02:57,033 --> 00:02:58,800
which we're going to call plt.

59
00:02:58,800 --> 00:03:01,800
That's the usual name
for this Pyplot module.

60
00:03:02,033 --> 00:03:02,400
All right.

61
00:03:02,400 --> 00:03:06,633
And finally we're going to import
a final library which is Defenders library

62
00:03:06,633 --> 00:03:11,400
a super useful library to preprocess
your data sets and mostly to import them.

63
00:03:11,666 --> 00:03:12,900
And same for this library.

64
00:03:12,900 --> 00:03:16,266
We're going to add the simple shortcut
name p d.

65
00:03:16,633 --> 00:03:18,866
And there you go. Congratulations.

66
00:03:18,866 --> 00:03:22,100
Now not only you know
how to import the libraries, but mostly

67
00:03:22,100 --> 00:03:25,800
you have your very first data
preprocessing tool in your toolkit.

68
00:03:26,266 --> 00:03:27,666
So now we're going to move on to

69
00:03:27,666 --> 00:03:31,166
the second tool
which will be to import the data set.

70
00:03:31,500 --> 00:03:33,900
And that's exactly what we'll
do in the next tutorial.