1
00:00:00,810 --> 00:00:08,080
In this video, we will see how to create decision tree that we have trained on our data.

2
00:00:09,090 --> 00:00:13,760
Now, plotting a busy uncreate is not an easy task in Python.

3
00:00:14,280 --> 00:00:15,990
It requires a lot of effort.

4
00:00:17,250 --> 00:00:20,730
We are not going into detail on how to grow it.

5
00:00:20,910 --> 00:00:27,450
We are just providing you code and we are just providing you the instructions on how to run it.

6
00:00:28,500 --> 00:00:32,040
So the first step is to create a dot file.

7
00:00:33,390 --> 00:00:37,680
Then we need to convert this dot file into an image.

8
00:00:38,250 --> 00:00:40,080
Then we will use this inmate.

9
00:00:41,170 --> 00:00:42,220
To create a graph.

10
00:00:43,630 --> 00:00:45,130
Now, this is the code.

11
00:00:46,510 --> 00:00:47,530
Remember this code?

12
00:00:47,920 --> 00:00:49,060
This is our variable.

13
00:00:50,730 --> 00:00:54,380
We need to use export, underscore Graff method.

14
00:00:54,930 --> 00:01:00,470
And here we have to provide over that aggressor KRI onto aggressor object.

15
00:01:01,380 --> 00:01:05,700
Since our object is rectory, we provided this as an argument.

16
00:01:07,840 --> 00:01:13,590
Now, in the next step, we have to import inmates from eye by standard display.

17
00:01:15,500 --> 00:01:21,590
We will use this image to display graph in our Biton notebook.

18
00:01:23,120 --> 00:01:33,350
The next step is to import by door, plus normally by default, this library is not installed in your

19
00:01:33,410 --> 00:01:34,450
anakonda packet.

20
00:01:34,880 --> 00:01:39,320
So first you have to install it to install this.

21
00:01:39,650 --> 00:01:43,100
You need to go to your and command prompt.

22
00:01:43,430 --> 00:01:49,540
And there you need to write KONDA Space install is space Bidart.

23
00:01:49,700 --> 00:02:00,260
Plus, if you are not using Anakonda, you have to go to your common ground and write up the space,

24
00:02:00,770 --> 00:02:02,770
install space by door.

25
00:02:02,970 --> 00:02:12,050
Plus, we have provided instructions on how to install Biodome plus in the description of this video.

26
00:02:12,530 --> 00:02:17,720
So check out if you want to plot your regression tree into Biton.

27
00:02:18,780 --> 00:02:29,850
Next, if you have in sorry pilot plus you can import Bidart plus and then use this Dort data that we

28
00:02:29,850 --> 00:02:33,330
have created earlier to create the regression tree.

29
00:02:35,100 --> 00:02:37,400
So we will run all this for commands.

30
00:02:43,920 --> 00:02:44,700
You can see.

31
00:02:45,710 --> 00:02:49,790
This is a word, this young tree here on the top.

32
00:02:50,630 --> 00:02:52,280
You can get the variable name.

33
00:02:53,860 --> 00:02:56,090
Year X squared, record.

34
00:02:56,130 --> 00:02:56,550
Three.

35
00:02:56,580 --> 00:02:58,260
This is our fourth variable.

36
00:02:58,350 --> 00:03:02,520
All four X data frame, which is budget.

37
00:03:03,920 --> 00:03:05,480
And here we have the condition.

38
00:03:06,080 --> 00:03:12,590
If the budget is less than thirty seven thousand, then it will follow this branch.

39
00:03:13,750 --> 00:03:20,230
And if the budget is greater than thirty seven thousand nine hundred eighty two, it will follow this

40
00:03:20,230 --> 00:03:20,670
branch.

41
00:03:22,400 --> 00:03:25,590
You can also see the average value of this symbol.

42
00:03:26,420 --> 00:03:31,680
So this is over Naude, the total number of observations is 404.

43
00:03:32,630 --> 00:03:39,710
And the average value of collection here is forty five thousand after the first split.

44
00:03:40,340 --> 00:03:47,690
So if your budget is more than this value, the average value for this symbol is seventy one thousand.

45
00:03:48,260 --> 00:03:52,430
So the average really of collection for this bucket is seventy one thousand.

46
00:03:52,760 --> 00:03:55,820
And the total number of samples in this book is 75.

47
00:03:56,630 --> 00:04:03,230
Whereas if you're budget value is less than thirty seven thousand nine hundred eighty two, the average

48
00:04:03,230 --> 00:04:05,760
value of your collection is thirty nine thousand.

49
00:04:06,020 --> 00:04:09,730
And the number of observations in this bucket is 392.

50
00:04:10,670 --> 00:04:15,730
And on the bottom, you can see we have eight leaf notes.

51
00:04:17,180 --> 00:04:20,660
And these are the average value of each leaf node.

52
00:04:21,080 --> 00:04:25,540
You can also get to the number of observation following and two, these leave notes.

53
00:04:27,130 --> 00:04:33,580
So for this leaf, our extra variable is less than thirty seven thousand nine hundred eighty two.

54
00:04:34,450 --> 00:04:37,990
Our X10 variable is less than 40000.

55
00:04:38,440 --> 00:04:44,410
Our first variable, that is Exito very well, is less than 135 and so on.

56
00:04:45,490 --> 00:04:48,770
So you can assign conditions to all this bucket.

57
00:04:49,270 --> 00:04:54,430
And these are the average values of collection in this bucket.

58
00:04:55,060 --> 00:04:58,660
And that is how you visualize CCRI and Biton.