1
00:00:00,600 --> 00:00:06,510
And this will do will create our first model using our extreme and wide-screen data.

2
00:00:07,840 --> 00:00:15,210
Will follow the same step as we do for every classification or regression model in Ashkelon.

3
00:00:16,440 --> 00:00:21,490
That is will first create our regulation or classifier object.

4
00:00:21,990 --> 00:00:25,390
Then we'll fit a word Ekstrand and Vytorin data.

5
00:00:25,500 --> 00:00:28,610
And to that classifier or regress that object.

6
00:00:29,310 --> 00:00:36,960
And then we will predict the word dependent variable values using our model and extreme and X test data.

7
00:00:38,730 --> 00:00:42,660
First, we need to import CCRI from Escalon.

8
00:00:44,940 --> 00:00:52,740
And then will create an object rectory here, rectory is our variable name with a sense for regression

9
00:00:52,740 --> 00:00:53,130
tree.

10
00:00:54,400 --> 00:00:58,210
And we lose Creedmore decision tree regrets at.

11
00:01:00,490 --> 00:01:00,870
Greed.

12
00:01:01,130 --> 00:01:03,880
Don't be angry, Degreaser is our function.

13
00:01:04,780 --> 00:01:07,090
And there are several parameters inside it.

14
00:01:09,750 --> 00:01:16,090
If you remember, we have discussed the terminologies of different dumps of Kree.

15
00:01:16,960 --> 00:01:20,440
You will find all those items as a parameter here.

16
00:01:21,550 --> 00:01:24,940
And if you remember what his maximum depth that is.

17
00:01:25,150 --> 00:01:28,500
Number of layers in your tree here.

18
00:01:28,780 --> 00:01:31,300
We don't want to overfit our tree.

19
00:01:31,450 --> 00:01:35,590
That's why we are keeping maximum depth equal to three.

20
00:01:35,680 --> 00:01:37,630
You can keep any other values.

21
00:01:38,350 --> 00:01:44,130
But to make your tree interpretable, don't accept this maximum depth.

22
00:01:44,320 --> 00:01:46,000
Values beyond five.

23
00:01:47,940 --> 00:01:51,740
We are running our first murder with maximum depth equal to three.

24
00:01:52,190 --> 00:01:54,710
You can choose any other values if you want.

25
00:01:58,750 --> 00:02:00,040
We will execute this.

26
00:02:01,260 --> 00:02:03,410
That green object is ready.

27
00:02:04,430 --> 00:02:07,140
Now our second step is Stopford extreme.

28
00:02:07,390 --> 00:02:11,080
And by train and to our direct gry object.

29
00:02:12,940 --> 00:02:17,470
We'll just use Rectory don't fit, don't fit, it is our method.

30
00:02:17,890 --> 00:02:21,160
And there are two parameters here, extreme and wide-screen.

31
00:02:25,470 --> 00:02:26,030
So what?

32
00:02:26,370 --> 00:02:26,750
Right.

33
00:02:27,010 --> 00:02:29,370
CCRI regulator object is ready.

34
00:02:29,430 --> 00:02:37,100
We can now use this as a treat to predict values of why using our x crane and expressed very well.

35
00:02:39,880 --> 00:02:42,790
No, predicting values is very easy.

36
00:02:43,680 --> 00:02:44,850
Will use victory.

37
00:02:45,160 --> 00:02:46,420
Don't predict method.

38
00:02:47,530 --> 00:02:53,080
And in this my third, we just have to provide our expertise wills, as you can see, we are saving

39
00:02:53,140 --> 00:02:54,980
our credit values, losing.

40
00:02:55,120 --> 00:02:59,380
Why underscore green, underscore pride and why underscore this?

41
00:02:59,410 --> 00:03:07,060
Underscore pride for our extranet and X has taught us it will just execute this.

42
00:03:07,060 --> 00:03:07,390
Come on.

43
00:03:09,760 --> 00:03:16,020
Now, we have created this Vikrant bread and wide spread, we will use this particular to a.

44
00:03:17,410 --> 00:03:20,000
To calculate the performance of our model.

45
00:03:22,020 --> 00:03:23,910
Now to see the projected values.

46
00:03:25,160 --> 00:03:26,090
You just have to, right?

47
00:03:26,120 --> 00:03:28,110
Why underscore tests, underscore pride.

48
00:03:29,120 --> 00:03:32,670
This will give you the trade we'll use on our test dataset.

49
00:03:35,020 --> 00:03:38,400
If you execute this are authentic values.

50
00:03:40,590 --> 00:03:45,870
Now, in the next lecture, we will use this predator to allow lose to calculate the performance of

51
00:03:45,870 --> 00:03:46,470
our modern.
