0
1
00:00:01,020 --> 00:00:07,440
The first big Python programming concept that we're going to talk about are variables. When you think
1

2
00:00:07,440 --> 00:00:10,200
about programming and software at a high level,
2

3
00:00:10,260 --> 00:00:12,630
it's all about manipulating data.
3

4
00:00:12,690 --> 00:00:18,960
For example, if you create a computer game, then you'll need to track the player score, the number of lives
4

5
00:00:18,960 --> 00:00:22,430
that the player has and what level the player is on.
5

6
00:00:22,470 --> 00:00:28,890
Of course, all this data changes over time as we play the video game and the program runs.
6

7
00:00:28,890 --> 00:00:36,400
So where do you as the programmer store this data and how do you track it? The Python programming language
7

8
00:00:36,400 --> 00:00:38,870
has a very convenient solution.
8

9
00:00:39,130 --> 00:00:44,830
You would store the game score in a variable. What is a variable?
9

10
00:00:45,140 --> 00:00:48,640
A variable is just a container for data.
10

11
00:00:48,640 --> 00:00:54,860
The important thing to understand about Python variables is that they both have a name and a value.
11

12
00:00:54,880 --> 00:01:02,920
For example, if I wanted to track my age inside a program I can simply create a variable called "myAge".
12

13
00:01:02,920 --> 00:01:12,520
"myAge" is the name of the variable, but because "myAge" is a container for data I can also give it a value,
13

14
00:01:12,550 --> 00:01:19,500
so I can stick the number 32 inside "myAge". In programming lingo
14

15
00:01:19,500 --> 00:01:27,460
this is called assigning a value to a variable. Now, as time passes and our program runs,
15

16
00:01:27,460 --> 00:01:34,810
we might need to update this variable, and we can do that simply by updating the value inside "myAge"
16

17
00:01:34,960 --> 00:01:44,980
with a new value. So we can replace the value 32 with the value 33. Now, the question is what would this
17

18
00:01:44,980 --> 00:01:52,230
look like in Python syntax? Inside the Python intro notebook that we created,
18

19
00:01:52,230 --> 00:01:57,650
we're going to create a variable called "myAge" and give it a value just like this"
19

20
00:01:57,750 --> 00:02:09,880
myAge = 32. This name "myAge" is now how we can refer to the number 32. To display the value inside
20

21
00:02:09,930 --> 00:02:12,340
the "myAge" variable below the cell,
21

22
00:02:12,340 --> 00:02:17,050
I'm simply going to write "print(myAge)".
22

23
00:02:19,750 --> 00:02:25,090
Hitting Shift+Enter on the keyboard will evaluate the code in the cell and then display this value
23

24
00:02:25,330 --> 00:02:28,810
contain inside the "myAge" variable.
24

25
00:02:28,810 --> 00:02:33,130
Now suppose our program is running and we need to update the value inside this variable.
25

26
00:02:33,250 --> 00:02:42,400
We can change this value of "myAge" simply by assigning a new value to it say with "myAge = 33".
26

27
00:02:42,460 --> 00:02:49,740
If I write "print(myAge)" now and hit Shift+Enter we'll see the value 33 printed out.
27

28
00:02:49,800 --> 00:02:51,860
No surprises there. Now,
28

29
00:02:51,910 --> 00:02:55,610
variables are super handy because you can do calculations with them.
29

30
00:02:55,680 --> 00:02:59,340
For example, if we write "print(myAge/3)"
30

31
00:03:03,670 --> 00:03:09,940
and hit Shift+Enter, we'll see 11.0 printed out below the cell.
31

32
00:03:09,940 --> 00:03:15,100
Also, we can take these two concepts that we just covered and take it up a step in complexity.
32

33
00:03:15,130 --> 00:03:19,260
Let's say we want to update the value of myAge like this.
33

34
00:03:19,390 --> 00:03:26,650
myAge = myAge + 1
34

35
00:03:26,710 --> 00:03:31,690
Now if it's your first time learning to program, this line of code is going to look very strange, but
35

36
00:03:31,690 --> 00:03:38,950
what's going on is that we're taking the previous value of myAge, namely the value 33 on the right hand
36

37
00:03:38,950 --> 00:03:44,490
side of the equals sign and then adding one to it. With that equals sign
37

38
00:03:44,500 --> 00:03:55,460
operator we are storing this new value - 34 (33 plus 1) inside of myAge on the left hand side.
38

39
00:03:55,480 --> 00:04:00,600
In other words, we are overwriting the previous value stored inside the
39

40
00:04:00,610 --> 00:04:13,320
myAge variable. If we write "print(myAge)" now and hit Shift+Enter, we'll see our value update to 34.
40

41
00:04:14,890 --> 00:04:22,260
So let's do a quick exercise on using and manipulating variables in Python. As a challenge,
41

42
00:04:22,480 --> 00:04:29,750
can you create a variable called "restaurantBill" and set its value equal to 36.17?
42

43
00:04:29,920 --> 00:04:37,690
Then create a variable called "serviceCharge" and set its value equal to 0.125. 12 and
43

44
00:04:37,690 --> 00:04:42,850
a half percent seems to be the standard rate that restaurants suggest as a tip in London these days.
44

45
00:04:44,020 --> 00:04:49,970
Finally, print out the amount of tip below the cell that you would need to add to the bill.
45

46
00:04:50,020 --> 00:04:52,120
I'll give you a few seconds to figure this out.
46

47
00:04:52,150 --> 00:04:55,900
So pause the video. Okay,
47

48
00:04:55,930 --> 00:04:59,170
so here's the solution. For part 1,
48

49
00:04:59,230 --> 00:05:06,650
we'll write "restaurantBill = 36.17". For part 2,
49

50
00:05:06,730 --> 00:05:08,870
we'll write
50

51
00:05:09,040 --> 00:05:15,680
"serviceCharge = 0.125"; and for part 3 we'll write
51

52
00:05:15,700 --> 00:05:18,500
print()
52

53
00:05:18,970 --> 00:05:20,670
restaurantBill
53

54
00:05:20,680 --> 00:05:22,570
Times serviceCharge.
54

55
00:05:26,200 --> 00:05:27,880
When I hit Shift+Enter,
55

56
00:05:27,910 --> 00:05:33,000
I'll see that the value of the tip is 4.52.
56

57
00:05:33,040 --> 00:05:38,470
Now all that's left to do is asking the waiter if the restaurant is pocketing the money, or if the tip
57

58
00:05:38,470 --> 00:05:41,270
really does go to the staff.
58

59
00:05:41,270 --> 00:05:48,190
Now, one thing that could happen to you at this point is that you've made a typo. In programming,
59

60
00:05:48,200 --> 00:05:50,340
everything is case sensitive.
60

61
00:05:50,540 --> 00:05:58,820
And if our variables names didn't match exactly to how they were defined, we're gonna get some unexpected
61

62
00:05:59,060 --> 00:06:00,590
errors.
62

63
00:06:00,620 --> 00:06:08,090
For example, if this capital B in my print statement instead was a lower case b and we had hit Shift+Enter
63

64
00:06:08,120 --> 00:06:15,010
we would have gotten the following error - "NameError: 'restaurantBill'
64

65
00:06:15,070 --> 00:06:21,880
is not defined", and this is because this restaurantbill and this restaurantBill are considered to
65

66
00:06:21,880 --> 00:06:24,390
be completely different entities.
66

67
00:06:24,790 --> 00:06:30,760
So you want to make sure that you never have any typos in your variable names.
67

68
00:06:30,760 --> 00:06:31,930
The same of course is true
68

69
00:06:31,960 --> 00:06:33,770
if we miss out a letter.
69

70
00:06:33,790 --> 00:06:42,010
So for example, if we wrote "seviceCharge" instead of "serviceCharge", then we would get exactly the same
70

71
00:06:42,040 --> 00:06:45,700
error - our variable is not defined.
71

72
00:06:45,760 --> 00:06:51,400
In other words, Python can't find something that matches this name.
72

73
00:06:51,400 --> 00:06:54,220
So let's put that "r" back where it belongs.
73

74
00:06:54,470 --> 00:07:00,550
Now that we've learned a thing or two about Python programming and variables, let's revisit the code
74

75
00:07:00,550 --> 00:07:08,400
that we wrote when we were estimating our movie revenue. Since I saved my previous work in a Python notebook
75

76
00:07:08,490 --> 00:07:11,030
and added it to the MLProjects folder,
76

77
00:07:11,100 --> 00:07:12,480
I'm just going to open that now.
77

78
00:07:15,790 --> 00:07:16,110
Now,
78

79
00:07:16,120 --> 00:07:21,240
even though this code is a little bit more complex than what we have written just now,
79

80
00:07:21,400 --> 00:07:25,890
can you spot the variables in this piece of code?
80

81
00:07:25,990 --> 00:07:27,720
Now there's actually quite a few.
81

82
00:07:27,940 --> 00:07:33,990
And you'll notice that the variables are often to the left hand side of an equal sign.
82

83
00:07:34,000 --> 00:07:44,900
So for example data is a variable, the capital X is a variable, the lower case y is also a variable.
83

84
00:07:44,900 --> 00:07:52,370
All of these variables are holding on to data, but in contrast the variables myAge and restaurantBill,
84

85
00:07:53,000 --> 00:07:59,750
these X and y variables are not holding on to a single value, they are holding onto lots and lots of
85

86
00:07:59,750 --> 00:08:01,950
values at the same time.
86

87
00:08:02,030 --> 00:08:06,950
And I'm going to show you guys how that works in a little bit. But before I do that,
87

88
00:08:06,960 --> 00:08:13,230
let's talk about another key concept that we need to understand when working with variables, namely data
88

89
00:08:13,230 --> 00:08:20,600
types. In machine learning and programming, more generally, we'll be working with different kinds of data.
89

90
00:08:20,750 --> 00:08:29,750
We'll be working with text, decimal numbers, tables of data, columns of data, images, sounds, video, all sorts.
90

91
00:08:31,480 --> 00:08:36,230
And a programming language like Python will categorize this data.
91

92
00:08:36,230 --> 00:08:42,650
In other words different kinds of data, like text or decimal numbers and whole numbers, will have a different
92

93
00:08:42,650 --> 00:08:43,840
category.
93

94
00:08:43,850 --> 00:08:48,320
In other words, they will have a different data type.
94

95
00:08:48,320 --> 00:08:53,600
Now you can think of data types like this children's toy where you have to fit the right shape into
95

96
00:08:53,600 --> 00:08:55,060
the hole.
96

97
00:08:55,070 --> 00:09:01,430
So for example when you try to put a decimal number somewhere where Python expects to have a string,
97

98
00:09:01,790 --> 00:09:04,350
you'll often find that you have a problem.
98

99
00:09:04,550 --> 00:09:07,840
And this makes sense when you think about it intuitively.
99

100
00:09:07,940 --> 00:09:10,980
So you have some Python code that makes a calculation.
100

101
00:09:11,090 --> 00:09:13,160
So you're adding two things together.
101

102
00:09:13,280 --> 00:09:18,600
If those two things are numbers then you're good; 5+10 is 15.
102

103
00:09:19,190 --> 00:09:26,290
But if one of those things is another kind of data, like, I don't know, a home address you have a problem.
103

104
00:09:26,360 --> 00:09:30,100
Your program will crash or it's going to do something very unexpected.
104

105
00:09:30,110 --> 00:09:35,690
Trying to evaluate 5 + 21 James Street.
105

106
00:09:35,690 --> 00:09:38,990
It doesn't make any sense. Now for the most part,
106

107
00:09:39,080 --> 00:09:45,470
Python will actually take care of the data types behind the scenes, so it's not something that's kind
107

108
00:09:45,470 --> 00:09:49,960
of at the forefront of the programming syntax of the Python code.
108

109
00:09:49,970 --> 00:09:56,360
But let me show you how you can actually see the data type in Jupyter notebook because we can ask Python
109

110
00:09:56,540 --> 00:10:04,670
what category something belongs to by writing "type()" and then putting something between those
110

111
00:10:04,670 --> 00:10:13,690
two parentheses. So for example, if I write "type(33)" and hit Shift+Enter, the type of this whole number is
111

112
00:10:13,910 --> 00:10:20,080
int. Int stands for integer, which makes sense.
112

113
00:10:20,440 --> 00:10:24,940
But there's also quite a few other types with maybe less intuitive names.
113

114
00:10:24,940 --> 00:10:33,800
So let me introduce a couple of them; if we write "type(33.6)" and hit Shift+Enter, we get
114

115
00:10:33,800 --> 00:10:38,240
to see that decimal numbers are classified differently.
115

116
00:10:38,240 --> 00:10:47,570
Decimal numbers are classified as floats or floating point numbers. Floating point numbers are the type
116

117
00:10:47,780 --> 00:10:53,430
that you will usually be working with every time you're dealing with numbers that have a decimal point.
117

118
00:10:53,430 --> 00:11:01,960
Now let me show you what the type for text is called so can write "type()" and then between the parentheses
118

119
00:11:02,260 --> 00:11:09,350
I'm going to open the single quotes and add my name. When I hit Shift+Enter,
119

120
00:11:09,460 --> 00:11:19,800
I can see that type for text is called "str", and "str" stands for string. The word string is just computer
120

121
00:11:19,800 --> 00:11:26,910
jargon for a piece of text or a sequence of characters. And the way that you can tell that something
121

122
00:11:27,000 --> 00:11:33,090
is considered to be a string is by the fact that strings are always, always surrounded by either single
122

123
00:11:33,420 --> 00:11:40,710
or double quotes and Jupyter notebook is actually also very helpful with the syntax highlighting so
123

124
00:11:40,710 --> 00:11:45,790
that the Python code that is considered to be a string actually has a different color.
124

125
00:11:45,810 --> 00:11:49,840
So in this case the string is marked as red.
125

126
00:11:49,860 --> 00:11:55,370
Let's take another look at the Python code that we wrote previously and see if we can spot a string.
126

127
00:11:55,550 --> 00:12:02,900
Here we can see that when we were writing "pd.read_csv()", the value
127

128
00:12:02,930 --> 00:12:08,120
that we specified between these two parentheses was a string.
128

129
00:12:08,120 --> 00:12:15,890
It was a piece of text marked by the single quotes. And the same is true for the column names - "production_budget_usd";
129

130
00:12:15,940 --> 00:12:22,940
and it's also true for the X label on our graph and all the other
130

131
00:12:22,940 --> 00:12:27,080
parts where the code is highlighted in red.
131

132
00:12:27,250 --> 00:12:29,490
Okay, so far so good.
132

133
00:12:29,590 --> 00:12:37,090
We've just introduced you to a whole bunch of new programming jargon - variables, data types, int for integers,
133

134
00:12:37,150 --> 00:12:42,530
float for decimal numbers, str for pieces of text or strings.
134

135
00:12:42,700 --> 00:12:45,880
But how do data types relate to variables?
135

136
00:12:45,880 --> 00:12:53,560
What's the connection between the two? Well, if the data has a certain type then so does the variable
136

137
00:12:53,620 --> 00:13:00,410
containing the data, because after all a variable is just a container. Now, for example, let's see what
137

138
00:13:00,410 --> 00:13:08,210
the data type is for the variable myAge so I can write type and then between the parentheses I'm going
138

139
00:13:08,210 --> 00:13:16,010
to supply myAge and hit Shift+Enter and here we can see that, because myAge holds on to whole numbers,
139

140
00:13:16,310 --> 00:13:22,500
the data type of the variable is also integer or int.
140

141
00:13:22,520 --> 00:13:26,150
Now let's take a look at the data type for restaurantBill,
141

142
00:13:29,390 --> 00:13:35,320
and I'm sure it's no surprise that, because restaurantBill is holding on to a decimal number 
142

143
00:13:35,330 --> 00:13:36,840
36.17,
143

144
00:13:36,840 --> 00:13:45,850
it's a type float. If I was to create a new variable called "myName" and set it equal to the string
144

145
00:13:45,870 --> 00:13:56,690
Philip, I can write "type(myName)", hit Shift+Enter and I can see that the myName variable is a string.
145

146
00:13:57,560 --> 00:14:03,890
Now let's take another look at the notebook where we ran our regression, and let me ask you this: what
146

147
00:14:03,890 --> 00:14:12,090
do you think is the data type of the variables data, capital X and lowercase y?
147

148
00:14:12,120 --> 00:14:18,350
This is the mystery that we're gonna be exploring over the next couple of lessons. I'll see you there.