1
00:00:00,420 --> 00:00:01,210
Welcome back.

2
00:00:01,590 --> 00:00:06,300
Well we've had a look at a few different ways of describing data that we've imported depend is in into

3
00:00:06,360 --> 00:00:07,630
a data frame.

4
00:00:07,650 --> 00:00:12,920
Let's have a look at a couple of different ways we can view and select data.

5
00:00:13,050 --> 00:00:15,770
So come down here we'll put it another little heading in.

6
00:00:15,870 --> 00:00:25,110
So ago viewing and selecting data we're going to hit escape an M for markdown shift and enter because

7
00:00:25,110 --> 00:00:29,320
we're keeping our notebooks nice and clean and communicative right from the beginning.

8
00:00:29,640 --> 00:00:36,240
So to begin one of the first things you'll do is have a look at the head of your data frame.

9
00:00:36,250 --> 00:00:38,360
Now remember this has brackets after it.

10
00:00:38,390 --> 00:00:39,810
So it's a function.

11
00:00:39,810 --> 00:00:45,980
What this is going to do is return the first or the top five rows of your data frame.

12
00:00:46,030 --> 00:00:47,870
Now you might be thinking Why five.

13
00:00:47,910 --> 00:00:49,650
Well that's a good question.

14
00:00:49,650 --> 00:00:54,100
I'm not entirely sure but five seems like a pretty good number.

15
00:00:54,120 --> 00:00:59,730
And so in practice what you'll probably be doing is manipulating your data frame in some way and then

16
00:00:59,730 --> 00:01:03,340
calling head on it fairly often so you don't have.

17
00:01:03,420 --> 00:01:05,850
Now our data frames only 10 rows.

18
00:01:05,850 --> 00:01:10,880
So calling the whole thing isn't actually too bad but imagine if you had thousands of these.

19
00:01:11,070 --> 00:01:17,490
So what head does is it gives you a quick snapshot in a small little space of what your data frame contains.

20
00:01:17,490 --> 00:01:22,700
So if you make a couple quick changes here and there you might want to look at the top five rows and

21
00:01:22,710 --> 00:01:24,080
a five isn't enough.

22
00:01:24,150 --> 00:01:30,150
Maybe you want to look at the top seven or the beautiful thing about head is that you can type in a

23
00:01:30,150 --> 00:01:33,420
number here and it will return that many rows.

24
00:01:33,720 --> 00:01:37,690
So you can have a play around and return whatever amount of rows you want.

25
00:01:37,710 --> 00:01:45,370
Now for some reason if you wanted the bottom of your data frame you can use dot tail and that will return.

26
00:01:45,510 --> 00:01:48,800
You might have guessed the bottom five rows.

27
00:01:48,800 --> 00:01:53,750
Now this is handy if you're doing some alterations on the bottom of your data frame rather than the

28
00:01:53,750 --> 00:01:54,440
top.

29
00:01:54,500 --> 00:01:57,200
And so the changes will only appear towards the bottom.

30
00:01:57,200 --> 00:02:02,260
Something like that entails much the same you can put in any kind of number you want here.

31
00:02:02,480 --> 00:02:07,350
Press shift and enter and it will turn the bottom three rows there.

32
00:02:07,360 --> 00:02:12,210
Now the next two functions that we're going to look at our lock and I lock.

33
00:02:12,210 --> 00:02:14,040
So let's write them down here.

34
00:02:14,080 --> 00:02:19,930
Lock and Dot I lock now before we go into the details.

35
00:02:20,040 --> 00:02:23,610
Remember where you always want to run code first.

36
00:02:23,610 --> 00:02:25,260
If in doubt run the code.

37
00:02:25,260 --> 00:02:26,600
That's our motto.

38
00:02:26,790 --> 00:02:32,970
So let's create a series here so we can demonstrate the difference between lock and lock so I'm going

39
00:02:32,970 --> 00:02:35,400
to make one called animals.

40
00:02:35,400 --> 00:02:36,530
What's your favorite animal.

41
00:02:36,600 --> 00:02:43,850
Maybe since we're using pandas maybe we'll put a panda in there and one more snake.

42
00:02:43,950 --> 00:02:46,130
Now we're going to set the index here.

43
00:02:46,440 --> 00:02:50,930
Now if we created a series by default the index will be 0 2.

44
00:02:50,940 --> 00:02:53,070
However many things we have here.

45
00:02:53,280 --> 00:03:00,090
But let's actually see that rather than talk about it because remember if in doubt run the code animals

46
00:03:00,330 --> 00:03:02,390
0 to 4 in order.

47
00:03:02,430 --> 00:03:09,310
But if we wanted our own custom index which we can do by passing the index parameter and then we put

48
00:03:09,310 --> 00:03:16,380
it in a list to say 0 3 we want this index to be out of order on purpose so we can demonstrate lock

49
00:03:16,530 --> 00:03:18,090
and I lock.

50
00:03:18,120 --> 00:03:18,740
There we go.

51
00:03:18,740 --> 00:03:20,670
So we have five things here.

52
00:03:20,820 --> 00:03:23,690
Five things here both Python lists.

53
00:03:23,760 --> 00:03:28,550
Now if we run this we can see that our index has changed.

54
00:03:29,160 --> 00:03:30,540
So let's have a look.

55
00:03:30,720 --> 00:03:34,140
Animals don't lock three.

56
00:03:34,260 --> 00:03:41,310
If we call this what do you think we'll come back and lock you can consider is short for location.

57
00:03:41,670 --> 00:03:47,040
So let's try shift into dog and Snake.

58
00:03:47,040 --> 00:03:48,070
Beautiful.

59
00:03:48,120 --> 00:03:52,860
So what lock refers to is the index numbers.

60
00:03:53,040 --> 00:03:58,920
So because we have three and three here it returns two items.

61
00:03:58,920 --> 00:04:04,090
Let's try animals not lock nine.

62
00:04:04,170 --> 00:04:05,610
What do you think this will come back with.

63
00:04:05,610 --> 00:04:10,070
Which animal will shift into bird.

64
00:04:10,100 --> 00:04:15,590
The reason being is because bird is at index nine so it created bird.

65
00:04:15,800 --> 00:04:18,050
One two three one two three.

66
00:04:18,050 --> 00:04:18,930
Index nine.

67
00:04:19,700 --> 00:04:24,070
Okay now let's try our car sales data frame.

68
00:04:24,100 --> 00:04:25,810
You're not sure what this looks like.

69
00:04:25,820 --> 00:04:28,260
We'll run it here so we can see.

70
00:04:28,280 --> 00:04:28,550
Okay.

71
00:04:28,690 --> 00:04:31,700
0 1 2 3 the index is still in order here.

72
00:04:31,700 --> 00:04:32,960
Beautiful.

73
00:04:32,960 --> 00:04:35,770
Now let's run a lock on this dot lock.

74
00:04:35,770 --> 00:04:37,640
3 What do you think this will come back with.

75
00:04:37,670 --> 00:04:40,920
Which car let's try it out.

76
00:04:42,170 --> 00:04:49,160
All beautiful comes back with position 3 a beautiful black BMW the twenty two thousand dollars can be

77
00:04:49,220 --> 00:04:55,570
all yours I could become a car salesman What do you think so now let's try another one let's try.

78
00:04:55,610 --> 00:05:05,490
I lock this time we'll put I lock here so got animals don't I lock three this time we're doing that

79
00:05:05,600 --> 00:05:09,420
I lock instead of just lock so we have five items here.

80
00:05:09,430 --> 00:05:11,600
What do you think this will come back with.

81
00:05:11,810 --> 00:05:13,010
Let's try it out.

82
00:05:13,010 --> 00:05:22,270
Shift into panda now so we can see it let's put a new cell here and go animals beautiful.

83
00:05:22,440 --> 00:05:34,800
So what I like refers to I like refers to position so we can see here 0 1 2 3 if this was in order remember

84
00:05:35,160 --> 00:05:41,810
Python lease and data frames in series start from zero I look refers to a position whereas the lock

85
00:05:41,820 --> 00:05:49,050
refers to index Let's try maybe with our car sales data frame car sales to I lock three

86
00:05:51,750 --> 00:05:59,680
it comes back with the same as lock that is because if we go up well actually let's just do it here.

87
00:05:59,790 --> 00:06:01,040
Car sales.

88
00:06:01,230 --> 00:06:08,300
If we have a look at our data frame I lock position three member because I like refers to position so

89
00:06:08,370 --> 00:06:13,680
zero one two three is the same as index three.

90
00:06:13,740 --> 00:06:20,570
So that's the main two points you have to remember about I lock and lock is that I lock refers to position

91
00:06:20,960 --> 00:06:31,590
and we'll put in here lock lock refers to index they're the main two differentiations there.

92
00:06:32,420 --> 00:06:32,900
OK.

93
00:06:33,330 --> 00:06:39,970
Now the beautiful thing about this is that with I lock and lock you can use slicing.

94
00:06:39,990 --> 00:06:44,880
So if you've ever used Python lists you might be familiar with slicing but if not I'll show you what

95
00:06:44,880 --> 00:06:49,570
that looks like might type in animals that I lock.

96
00:06:49,630 --> 00:07:00,300
Colon 3 and what this means is rather than talk about it let's run the code is give us the items in

97
00:07:00,300 --> 00:07:03,680
animals up to position 3.

98
00:07:03,720 --> 00:07:06,030
Now I lock doesn't include position 3.

99
00:07:06,510 --> 00:07:12,570
So give us remember stuff from 0 0 1 2 3.

100
00:07:12,690 --> 00:07:15,180
So it's going to give us up to 3.

101
00:07:15,180 --> 00:07:18,950
Now let's try it with the car sales data from car sales.

102
00:07:19,080 --> 00:07:20,490
Don't lock.

103
00:07:20,490 --> 00:07:23,690
Colon three shift and enter.

104
00:07:23,770 --> 00:07:24,490
There we go.

105
00:07:24,490 --> 00:07:28,590
It's given us up to and including index 3.

106
00:07:28,630 --> 00:07:40,310
Basically this is the same as calling head but with for car sales don't head for excellent.

107
00:07:40,370 --> 00:07:40,890
Okay.

108
00:07:41,010 --> 00:07:41,860
Let's keep going.

109
00:07:42,600 --> 00:07:48,060
Now we've seen this previously but let's say we wanted to select just this make column and have a look

110
00:07:48,060 --> 00:07:48,870
at what's going on there.

111
00:07:49,560 --> 00:07:56,100
So the way to select a column is to type in its name in square brackets next to the name of the data

112
00:07:56,100 --> 00:08:00,740
frame as a strength shift into there's the make column.

113
00:08:00,900 --> 00:08:02,700
Let's have a look at the color column.

114
00:08:02,910 --> 00:08:08,470
Car sales color Excellent.

115
00:08:08,480 --> 00:08:09,830
We've selected that.

116
00:08:09,830 --> 00:08:14,660
Now there's two ways you might see I want you to be familiar with both of these ways of selecting a

117
00:08:14,660 --> 00:08:15,080
column.

118
00:08:15,080 --> 00:08:15,510
So if we go.

119
00:08:15,510 --> 00:08:16,200
Car sales.

120
00:08:16,190 --> 00:08:20,980
Dot make it easy as it's hype then something like that.

121
00:08:20,990 --> 00:08:22,370
But this is the same thing.

122
00:08:22,730 --> 00:08:24,490
Let me demonstrate.

123
00:08:24,530 --> 00:08:31,310
Car sales make these two lines of code are going to do the exact same thing.

124
00:08:31,400 --> 00:08:32,480
Let's prove it.

125
00:08:32,510 --> 00:08:38,610
I'll put a split in here by hitting control shift minus any I make.

126
00:08:38,690 --> 00:08:39,860
Excellent.

127
00:08:39,870 --> 00:08:48,020
We're going to a car sales don't make now the only real difference between these two is the syntax.

128
00:08:48,020 --> 00:08:49,810
So whichever one you prefer.

129
00:08:49,880 --> 00:08:56,570
You might want to use but what you should know about the dot notation is that if your column name has

130
00:08:56,570 --> 00:09:00,190
a space in it the dot notation won't work.

131
00:09:00,230 --> 00:09:08,030
So if we tried to do this for odometer K M it will come back with an error.

132
00:09:08,630 --> 00:09:19,800
But if we put a cell in here and go car sales odometer K and run this it'll return the odometer.

133
00:09:19,850 --> 00:09:26,260
So there are two main differences between selecting a single column and now one last thing before this

134
00:09:26,260 --> 00:09:31,300
video gets too long is that if you wanted to select a single column but put a little bit of a filter

135
00:09:31,300 --> 00:09:33,370
on it let's start with the make column.

136
00:09:33,370 --> 00:09:38,600
But we only want rows with that car maker Toyota.

137
00:09:39,490 --> 00:09:42,340
So can you decipher what's going on here.

138
00:09:42,340 --> 00:09:43,980
We're saying car sales.

139
00:09:44,110 --> 00:09:44,500
Okay.

140
00:09:44,500 --> 00:09:45,550
And then we're passing it.

141
00:09:45,550 --> 00:09:47,020
This little condition here.

142
00:09:47,020 --> 00:09:49,340
This is called boolean indexing.

143
00:09:49,390 --> 00:09:51,770
Let's hit shift entered to have a look.

144
00:09:51,780 --> 00:09:58,410
So this is going to say hey panders give us a car sales data frame and this condition is I only want

145
00:09:58,410 --> 00:10:04,370
the car sales data in the make column which are equal to Toyota.

146
00:10:04,380 --> 00:10:12,480
Now what if we only wanted maybe those with over 100000 kilometers on the data on the the data on the

147
00:10:12,480 --> 00:10:13,450
odometer.

148
00:10:13,450 --> 00:10:19,980
We've got data on the brain as well we're saying it let's get odometer here Cam because remember we

149
00:10:19,980 --> 00:10:26,130
can't do the dot notation so you can create a habit which when you want to type usually you pick one

150
00:10:26,220 --> 00:10:30,300
way of selecting a single column and stick with that throughout your entire notebook.

151
00:10:30,510 --> 00:10:35,240
So we want greater than 100000.

152
00:10:35,260 --> 00:10:43,220
Now this is gonna give back the rows that fulfill this condition so you can imagine how advanced you

153
00:10:43,220 --> 00:10:47,990
could get with this if you had say 100000 different cars you might put in a few more different criteria

154
00:10:47,990 --> 00:10:54,080
in here but this is just a few simple use cases of how we can view and select data.

155
00:10:54,080 --> 00:10:57,710
Now there's a couple more we're going to go through but to prevent this video from going for too long

156
00:10:57,710 --> 00:10:59,530
we're going to take a little break.

157
00:10:59,660 --> 00:11:04,460
Go back over what we've gone just through and practice a little bit before the next one but otherwise

158
00:11:04,490 --> 00:11:07,700
we'll see some more ways of selecting and viewing data in the next video.