1
00:00:00,950 --> 00:00:05,670
Hello, all in all of a previous session, we have analyzed as much as we can.

2
00:00:05,690 --> 00:00:11,440
We have understand our data and in the previous session, we have applied a concept of what exactly

3
00:00:11,450 --> 00:00:18,530
the correlation and using correlation we have out, what are my topmost features that are going to contribute

4
00:00:18,530 --> 00:00:20,430
more to my machine learning model.

5
00:00:20,720 --> 00:00:26,610
After that, we have used in some basic human knowledge using some basic sense of humor.

6
00:00:26,630 --> 00:00:28,820
I just exclude some features.

7
00:00:29,060 --> 00:00:35,030
And after what we have to do in this session, this is exactly the very first statement for this particular

8
00:00:35,030 --> 00:00:35,480
session.

9
00:00:35,480 --> 00:00:39,710
We have to extract some derived features from data.

10
00:00:40,010 --> 00:00:42,660
So let me show you what exactly the did feature.

11
00:00:42,860 --> 00:00:48,890
So let me show you what exactly the data frame on which you have to find this feature.

12
00:00:49,280 --> 00:00:56,380
So in this data, I have to just pass my cat and a features and let's say this is exactly my data on

13
00:00:56,390 --> 00:00:56,960
a score.

14
00:00:57,260 --> 00:00:59,510
Get data from just executed.

15
00:00:59,540 --> 00:01:07,250
Now, if I am going to call my had all the data on the score, cat not had to get a preview.

16
00:01:07,250 --> 00:01:10,270
How my data looks like you will see over here.

17
00:01:10,280 --> 00:01:15,290
These are all the features on which I have to perform.

18
00:01:15,290 --> 00:01:16,810
Lots of preprocessing.

19
00:01:16,850 --> 00:01:23,000
You will see these are exactly the status and escorted from this feature.

20
00:01:23,030 --> 00:01:29,150
You have to extract your desired features, so you have to extract your you have to extract money.

21
00:01:29,450 --> 00:01:35,450
You have to extract because whenever you are going to build a machine learning model, that model will

22
00:01:35,450 --> 00:01:38,530
now understand what exactly this indeed.

23
00:01:38,720 --> 00:01:40,810
So you have to do in machine learning model.

24
00:01:41,010 --> 00:01:42,370
Yeah, this is my ear.

25
00:01:42,390 --> 00:01:43,250
This is my money.

26
00:01:43,370 --> 00:01:44,150
This is my date.

27
00:01:44,510 --> 00:01:49,310
So these are exactly all your desired features and you have to fetch from this data.

28
00:01:49,760 --> 00:01:51,710
So very first, what I have to do.

29
00:01:51,950 --> 00:01:55,690
Let me check what exactly the data of each and every feature.

30
00:01:55,700 --> 00:02:03,670
So I'm going to say it, cat dot data detector here, just like the cell and you will figure out this

31
00:02:03,680 --> 00:02:08,470
status date, find us by default, assign it as objected.

32
00:02:08,960 --> 00:02:11,690
But we know it supports some data from it.

33
00:02:11,960 --> 00:02:15,830
It means we have to convert this into some data format for this.

34
00:02:15,830 --> 00:02:18,760
I'm just going to say just call this to escort data.

35
00:02:18,770 --> 00:02:24,650
And here I'm going to say data and go get which is exactly this one.

36
00:02:24,950 --> 00:02:29,540
And on this, I have to exit this reservation and status and it's coded.

37
00:02:29,900 --> 00:02:34,580
And now what I have to do, I have to simply update this feature as well.

38
00:02:34,590 --> 00:02:42,090
So I'm going to say Reser version underscore is status and a code named after doing all this stuff.

39
00:02:42,150 --> 00:02:44,990
What I have to do, I have to just agree to this set.

40
00:02:45,380 --> 00:02:47,180
He will see you have some warnings.

41
00:02:47,480 --> 00:02:49,130
Just don't worry at all.

42
00:02:49,130 --> 00:02:54,200
These are just on account of some weather issues, some liability issues.

43
00:02:54,440 --> 00:02:56,930
Some of you might have some issues.

44
00:02:57,200 --> 00:02:58,370
There are lots of issues.

45
00:02:58,370 --> 00:03:04,730
So if you want to ignore or if you want to get rid of this warnings, you can import your learning module

46
00:03:04,730 --> 00:03:05,040
as well.

47
00:03:05,060 --> 00:03:07,010
So I would say import my warnings.

48
00:03:07,250 --> 00:03:14,450
And from this warning, what I have to do, I have to import something known as filter warning.

49
00:03:14,450 --> 00:03:17,900
Just best stab at in this filter warning.

50
00:03:17,900 --> 00:03:23,000
What I have to say, I have to say wherever warning will occur, just ignore this.

51
00:03:23,040 --> 00:03:26,900
That's what this blocks of code will do if I'm going to execute it.

52
00:03:26,900 --> 00:03:29,270
Now, you don't have any warning now.

53
00:03:29,480 --> 00:03:36,200
Your main task is you have to fetch your money and day from this data.

54
00:03:36,210 --> 00:03:38,150
Get off this, this, this.

55
00:03:38,420 --> 00:03:48,530
So to fetch your ear, you have to just collar d dot your over it now to fetch your munt you have to

56
00:03:48,530 --> 00:03:56,420
just call this D dot over it now to fetch your day you have to just call Dot.

57
00:03:56,600 --> 00:03:57,170
That's it.

58
00:03:57,560 --> 00:04:02,400
And let's say whatever it will return me, I have to store it somewhere else.

59
00:04:02,400 --> 00:04:08,930
So I'm going to say data on this go get and where you have to store it.

60
00:04:08,930 --> 00:04:13,130
Yeah I have to basically store it in a new feature so I have to define it.

61
00:04:13,130 --> 00:04:19,100
So using this data to get out of here, you can simply define that feature, that input.

62
00:04:19,430 --> 00:04:26,260
And after it, I have to define a new feature as data underscore, get all select statement.

63
00:04:26,330 --> 00:04:30,140
I'm going to say it is my month after doing all this stuff.

64
00:04:30,350 --> 00:04:36,980
I have to again access this data or get over there and hear what I have to do.

65
00:04:36,980 --> 00:04:40,790
I have to say this feature name or column name is nothing like to do.

66
00:04:41,360 --> 00:04:43,540
So just execute all this stuff.

67
00:04:43,550 --> 00:04:49,180
And again, if I'm going to access this data, underscore or get around this, if I'm going to call

68
00:04:49,640 --> 00:04:57,500
lot or do now, you will see you have three columns added the very first to the year, your month,

69
00:04:57,770 --> 00:05:00,200
your day, and now you will see over here.

70
00:05:00,540 --> 00:05:07,710
This feature, which is exactly this generation is in a school cadet, makes no sense at all because

71
00:05:07,710 --> 00:05:10,930
whatever you want to fetch from this, you have already fetch.

72
00:05:11,220 --> 00:05:13,410
So now you can simply drop this.

73
00:05:13,420 --> 00:05:19,290
I'm going to say to underscore cat dot drop here you have a function.

74
00:05:19,290 --> 00:05:21,900
And what I have to do if I drop this feature, that's it.

75
00:05:22,290 --> 00:05:28,220
Then I have to say to my actors, it goes to one because I had to drop in vertical way.

76
00:05:28,290 --> 00:05:30,200
Then I have to update my little frame as well.

77
00:05:30,210 --> 00:05:34,500
So I'm going to stay in place to just execute.

78
00:05:34,500 --> 00:05:36,220
All this stuff gets executed.

79
00:05:36,780 --> 00:05:42,400
Let's say I have to insert a new column in this data underscore CAT.

80
00:05:42,750 --> 00:05:51,570
Its name is cancellation because I didn't get this exactly that data frame where we have all the categorical

81
00:05:51,570 --> 00:05:54,960
features and we all lose this data.

82
00:05:54,980 --> 00:05:55,420
Let's go.

83
00:05:55,460 --> 00:06:00,600
Let me show you a thing which is exactly DataDot, these types.

84
00:06:00,650 --> 00:06:07,050
So if I'm going to call this Daibes, you will see a feature, which is exactly this is cancer, which

85
00:06:07,050 --> 00:06:08,730
this bond does assign.

86
00:06:08,730 --> 00:06:12,420
It has in sixty four, but we all lose.

87
00:06:12,690 --> 00:06:18,270
It is exactly an object because the values are exactly in the form of zeros and ones.

88
00:06:18,270 --> 00:06:24,180
So zero basically means your bookings are not going to cancel and one means your booking are going to

89
00:06:24,180 --> 00:06:24,590
cancel.

90
00:06:24,600 --> 00:06:27,920
It means it's objective to type column.

91
00:06:28,200 --> 00:06:32,790
So you can definitely consider this feature in this data set.

92
00:06:33,030 --> 00:06:36,470
So I'm going to say data underscore that.

93
00:06:36,810 --> 00:06:40,710
And here I could just define a new column like, say, cancellation.

94
00:06:40,740 --> 00:06:47,550
Actually, this is exactly my column and now I'm going to say data on a out of here.

95
00:06:47,550 --> 00:06:54,930
I have to say data of is underscore cancel all data of IZAN score, cancel it now.

96
00:06:54,940 --> 00:06:56,690
Just execute all this stuff.

97
00:06:56,940 --> 00:07:04,560
And if again I'm going to call my head over there now, you will figure out here you have a new feature

98
00:07:04,560 --> 00:07:06,450
which is exactly cancellation.

99
00:07:06,540 --> 00:07:08,400
So that's all about the session.

100
00:07:08,400 --> 00:07:14,640
In the upcoming session, we are going to perform our feature encoding techniques on our data because

101
00:07:14,880 --> 00:07:17,040
these are exactly my string data.

102
00:07:17,040 --> 00:07:21,300
You will see that at this resort hotel, no deposit.

103
00:07:21,330 --> 00:07:25,500
These all these features are exactly matching data and machine learning.

104
00:07:25,770 --> 00:07:30,390
Now, understand, this is string guitar, so you have to tell machine learning.

105
00:07:31,170 --> 00:07:37,890
It means you have to convert these string data into some Intisar format using some feature encoding

106
00:07:37,890 --> 00:07:40,200
techniques in the upcoming session.

107
00:07:40,210 --> 00:07:42,450
We are going to deal with all these techniques.

108
00:07:42,480 --> 00:07:44,550
So I hope you will love this session very much.

109
00:07:44,730 --> 00:07:45,420
Thank you.

110
00:07:45,480 --> 00:07:46,350
Have a nice day.

111
00:07:46,470 --> 00:07:47,310
Keep learning.

112
00:07:47,310 --> 00:07:48,180
Keep doing.

113
00:07:48,180 --> 00:07:49,050
Keep practicing.