1
00:00:00,200 --> 00:00:01,133
Hello my friends.

2
00:00:01,133 --> 00:00:04,333
Welcome to part one Data Preprocessing.

3
00:00:04,333 --> 00:00:08,033
And mostly welcome to the practical
activities of this course.

4
00:00:08,333 --> 00:00:11,966
My name is Adam
and this and I am more than happy

5
00:00:11,966 --> 00:00:13,800
to welcome you into this course.

6
00:00:13,800 --> 00:00:16,766
We're about to embark
on a super huge journey,

7
00:00:16,766 --> 00:00:20,400
but yet a very exciting one
because we're going to build together

8
00:00:20,400 --> 00:00:25,300
a series of machine learning models
from all the branches of machine learning.

9
00:00:25,300 --> 00:00:29,400
And speaking of these branches,
will you have them in front of you?

10
00:00:29,666 --> 00:00:33,433
That's the whole folder of this machine
learning it is it course

11
00:00:33,633 --> 00:00:37,433
structured into all the different
branches of machine learning,

12
00:00:37,566 --> 00:00:41,433
from data preprocessing,
regression, classification, clustering,

13
00:00:41,633 --> 00:00:45,300
association rule learning, reinforcement
learning and NLP and deep learning.

14
00:00:45,633 --> 00:00:49,533
And now we're about to start part
one day data preprocessing.

15
00:00:49,666 --> 00:00:52,800
And this is the first
very important step of this journey.

16
00:00:52,966 --> 00:00:56,500
Because indeed any time
you build a machine learning model,

17
00:00:56,600 --> 00:01:00,633
you always have a data
preprocessing phase to work on, right?

18
00:01:00,633 --> 00:01:05,200
You have to preprocess the data
in the right way, so that the machine

19
00:01:05,200 --> 00:01:09,400
learning model that you're going to build
can be trained the right way on the data.

20
00:01:09,700 --> 00:01:11,833
So we have to start with this step.

21
00:01:11,833 --> 00:01:15,600
This will not be the most exciting step,
but I will make sure to

22
00:01:15,633 --> 00:01:19,033
do this very efficiently
so that we can rapidly proceed

23
00:01:19,066 --> 00:01:23,166
to the branches of machine learning, where
we're going to build indeed some models.

24
00:01:23,333 --> 00:01:26,733
But before that,
we have to master data preprocessing.

25
00:01:26,966 --> 00:01:29,966
And that's exactly
what we're going to do in this part one.

26
00:01:30,266 --> 00:01:30,766
All right.

27
00:01:30,766 --> 00:01:31,966
So first of all,

28
00:01:31,966 --> 00:01:35,733
make sure that you have the same
as what I have on my computer right now.

29
00:01:36,000 --> 00:01:40,500
This is the whole machine learning is
that codes and data sets folder containing

30
00:01:40,500 --> 00:01:44,433
old Python codes and all the R codes,
as well as all the data sets

31
00:01:44,766 --> 00:01:48,333
in each of the practical activities
we're going to do together.

32
00:01:48,600 --> 00:01:52,400
The link of this folder was provided
to you in the previous section.

33
00:01:52,400 --> 00:01:53,633
You know, section one.

34
00:01:53,633 --> 00:01:56,166
So if we are on the same page
let's do this.

35
00:01:56,166 --> 00:02:00,000
Let's go inside this part1
data preprocessing folder

36
00:02:00,266 --> 00:02:02,233
which is structured the following way.

37
00:02:02,233 --> 00:02:06,300
It has only one section, luckily for us,
so that we can quickly get to the machine

38
00:02:06,300 --> 00:02:10,166
learning models
and as any folder in this course,

39
00:02:10,166 --> 00:02:13,166
well, it is structured
the following way in two folders.

40
00:02:13,166 --> 00:02:15,000
First Python and then R.

41
00:02:15,000 --> 00:02:18,833
Despite and folder contains
of course all the Python implementations

42
00:02:18,833 --> 00:02:22,700
and the data set, and this folder contains
all the R implementations

43
00:02:22,700 --> 00:02:24,100
with the same data set.

44
00:02:24,100 --> 00:02:28,633
So now we're going to go into Python
because each time in in each section,

45
00:02:28,633 --> 00:02:33,000
anytime we build a model we're going to
start first with Python and then R okay.

46
00:02:33,000 --> 00:02:36,800
So that if you are more interested
in Python, will you take the first part.

47
00:02:36,933 --> 00:02:39,766
And if you're more interested in R
you take the second part.

48
00:02:39,766 --> 00:02:42,533
And here
I want to say something very important.

49
00:02:42,533 --> 00:02:46,133
This course is not meant for you to master
both Python and R.

50
00:02:46,200 --> 00:02:50,066
We just covered the two tools
so that anyone can learn machine

51
00:02:50,066 --> 00:02:52,000
learning on their preferred tool.

52
00:02:52,000 --> 00:02:54,433
So if you prefer Python,
you can just do Python.

53
00:02:54,433 --> 00:02:56,266
If you prefer R, you can just to R.

54
00:02:56,266 --> 00:02:58,800
And if you want to learn both
well you're welcome to learn both.

55
00:02:58,800 --> 00:03:01,766
But you don't have to.
That's what I want to say okay.

56
00:03:01,766 --> 00:03:04,100
So we're going to start with Python here

57
00:03:04,100 --> 00:03:07,300
I'm going to teach you
all the tools on data preprocessing.

58
00:03:07,500 --> 00:03:08,233
And there you go.

59
00:03:08,233 --> 00:03:11,400
That's the files you will find
in this data preprocessing folder.

60
00:03:11,400 --> 00:03:15,333
In the Python folder you will first
find this data preprocessing tools.

61
00:03:15,333 --> 00:03:18,766
Implement version
which contains all the different tools

62
00:03:19,033 --> 00:03:23,400
of data preprocessing that
you might have to use on your data sets

63
00:03:23,400 --> 00:03:27,000
in order to preprocess them the right way
for your machine learning model.

64
00:03:27,400 --> 00:03:31,733
Then we have the data preprocessing
template, which will be very useful

65
00:03:31,733 --> 00:03:35,933
for us to tackle any data
preprocessing phase of our future.

66
00:03:35,933 --> 00:03:38,933
Machine learning model implementations
in the following parts.

67
00:03:39,000 --> 00:03:40,166
So you will see.

68
00:03:40,166 --> 00:03:42,533
You will absolutely love this template.

69
00:03:42,533 --> 00:03:46,166
And we also have this data in a CSV file,

70
00:03:46,233 --> 00:03:50,000
data dot CSV,
which is the data set on which

71
00:03:50,000 --> 00:03:53,700
I will show you how to implement
all these data preprocessing tools.

72
00:03:53,900 --> 00:03:56,833
And just to give you some context,
let's say that this data

73
00:03:56,833 --> 00:04:00,733
set belongs to a retail company
that collected some data

74
00:04:00,733 --> 00:04:04,900
from their customers, whether or not
they purchase a certain product.

75
00:04:04,900 --> 00:04:08,566
So here each of the rows correspond
to different customers.

76
00:04:08,566 --> 00:04:12,566
And for each of these customers, well,
this company gathered their country,

77
00:04:12,666 --> 00:04:17,166
their age, their salary and whether or not
they purchased their product.

78
00:04:17,166 --> 00:04:19,366
All right. So that's a simple data set.

79
00:04:19,366 --> 00:04:22,733
But I wanted to use a simple one
so that we can really focus

80
00:04:22,966 --> 00:04:25,766
on all the tools
we're going to learn in this section.

81
00:04:25,766 --> 00:04:31,566
And speaking of them, well, now I suggest
that we start this data preprocessing

82
00:04:31,600 --> 00:04:36,000
tools implementation because indeed this
course is really action based.

83
00:04:36,000 --> 00:04:39,000
You know, you're going to learn
by doing in this course.

84
00:04:39,000 --> 00:04:42,233
And therefore for this implementation
and each future

85
00:04:42,233 --> 00:04:45,866
implementation,
we will re-implement it from scratch.

86
00:04:45,966 --> 00:04:46,800
And so there you go.

87
00:04:46,800 --> 00:04:50,966
Our first implementation will be
for all the data preprocessing tools.

88
00:04:50,966 --> 00:04:53,200
And now we're going to open this file.

89
00:04:53,200 --> 00:04:57,333
And as I explained in the first section
of this course you can either open it

90
00:04:57,333 --> 00:05:02,500
in Google Colaboratory by just double
clicking here or on Jupyter Notebook.

91
00:05:02,500 --> 00:05:04,500
If you don't like Google Colaboratory.

92
00:05:04,500 --> 00:05:07,500
So feel free to select
your most comfortable environment.

93
00:05:07,533 --> 00:05:10,600
And for Google colaboratory lovers, well,
there we go.

94
00:05:10,600 --> 00:05:13,600
Let's open our data preprocessing tools.

95
00:05:13,866 --> 00:05:18,300
And first I'm going to show you exactly
the tools that we're going to implement.

96
00:05:18,300 --> 00:05:21,500
And mostly that you're going to learn
for your machine learning models.