1
00:00:00,390 --> 00:00:04,930
Hello and welcome back to our second structured data project.

2
00:00:05,070 --> 00:00:10,490
Now this one's really exciting right because we're going to be predicting the sale price of bulldozers.

3
00:00:10,530 --> 00:00:14,880
Who doesn't love bulldozer remember when I was a kid I used to have all these tractors like tanker trucks

4
00:00:15,240 --> 00:00:17,310
and used to play with them all the time.

5
00:00:17,400 --> 00:00:18,940
And this one's a little bit different.

6
00:00:19,020 --> 00:00:24,900
Of course it's a regression problem because we're trying to predict a number a.k.a. the sale price of

7
00:00:24,900 --> 00:00:25,720
bulldozers.

8
00:00:25,830 --> 00:00:30,230
But what we're gonna do is we're going to take data and in this case the data.

9
00:00:30,330 --> 00:00:32,080
See this little clock emoji.

10
00:00:32,160 --> 00:00:34,410
It has a time component to it.

11
00:00:34,440 --> 00:00:40,110
So that means in our previous project there was no time series it was just a big static data set whereas

12
00:00:40,110 --> 00:00:46,560
this one there's a sale date component which means that a bulldozer has been sold on a certain date.

13
00:00:46,590 --> 00:00:54,000
And so what we'll be doing is using sales from the past to try and predict sales in the future.

14
00:00:54,260 --> 00:00:59,910
And that's our problem definition right is we'll take our data which is full of bulldozer sales we're

15
00:00:59,910 --> 00:01:01,830
trying to predict the sale price.

16
00:01:01,830 --> 00:01:06,620
And I mean you might be looking at this emerging going that's a tractor Daniel that's not a bulldozer

17
00:01:06,660 --> 00:01:10,770
and you know there there's no bulldozer emoji sadly yet.

18
00:01:10,890 --> 00:01:12,900
That may be an update coming from Apple but who knows.

19
00:01:12,900 --> 00:01:14,600
Fingers crossed.

20
00:01:14,790 --> 00:01:18,610
Of course we're going to use our six step machine learning framework.

21
00:01:18,690 --> 00:01:24,090
We'll start by defining our problem which is of course seeing if we can predict the sale price of bulldozers

22
00:01:24,450 --> 00:01:29,580
we'll have a look at some data we'll find and an evaluation metric which is suitable for our problem.

23
00:01:29,670 --> 00:01:32,010
Well then have a look at the features of the data.

24
00:01:32,100 --> 00:01:36,240
We'll of course try some machine learning modelling because after all that's our goal is to build a

25
00:01:36,240 --> 00:01:42,030
machine learning model that helps us to breed evict the sale price of bulldozers and throughout the

26
00:01:42,210 --> 00:01:49,350
entire session we'll be doing experiments now we've seen what tools we can use and we're going to be

27
00:01:49,350 --> 00:01:53,860
using the exact same tools set that we used in the previous structured data project.

28
00:01:53,940 --> 00:01:58,800
So we'll have Panda's map plot layer but none imply for data analysis we'll be working within the Jupiter

29
00:01:58,800 --> 00:02:04,710
notebook using our Condor environment and we'll use psychic learn for any machine learning modelling

30
00:02:06,510 --> 00:02:12,030
now because this is a new project will approach it as if it's a new project.

31
00:02:12,240 --> 00:02:14,760
We've got our steps to take in a new project here.

32
00:02:14,910 --> 00:02:19,990
But we've been through these we've got our computer we've download installed mini Condor.

33
00:02:20,010 --> 00:02:22,250
This is where we're at now we're starting a new project.

34
00:02:22,740 --> 00:02:29,460
So we'll create a new project folder and now of course because we're using the exact same tools as what

35
00:02:29,460 --> 00:02:35,160
we did in the previous structured data project you could complete this entire project.

36
00:02:35,160 --> 00:02:39,480
The bulldozer sale price prediction in that previous folder.

37
00:02:39,570 --> 00:02:45,300
However for completeness we're going to create a new project folder or download our data from Kaggle.

38
00:02:45,330 --> 00:02:46,700
We'll get to that in a second.

39
00:02:46,950 --> 00:02:52,740
We'll create an environment which holds all of our tools that we'll be using we'll use our Jupiter notebook

40
00:02:52,770 --> 00:02:59,100
as our workspace to perform data analysis and manipulation using our name pi panders and map plot lib

41
00:02:59,100 --> 00:03:04,100
tools and then machine learning using psychic learn.

42
00:03:04,160 --> 00:03:09,560
Now of course if you do get stuck and don't worry you probably will I get stuck all the time.

43
00:03:09,560 --> 00:03:14,960
The steps that you can take is to follow along with the code as best you can try it for yourself.

44
00:03:14,960 --> 00:03:20,660
Remember our motto if in doubt run the code then if you want to find out about more about what a function

45
00:03:20,660 --> 00:03:21,380
is doing.

46
00:03:21,440 --> 00:03:26,330
Remember to press shift and tab to read the dock string which will give you some information of water

47
00:03:26,570 --> 00:03:28,050
what a function is doing.

48
00:03:28,190 --> 00:03:33,170
If you're still stuck then of course don't be afraid to search for it right you'll come across resources

49
00:03:33,170 --> 00:03:40,910
like stack overflow of the documentation for cyclone panders and then if you search for it and you find

50
00:03:40,910 --> 00:03:46,470
some information maybe you copy it from there from the documentation into your Jupiter notebook.

51
00:03:46,700 --> 00:03:53,600
Then of course try again remember if in doubt run the code and if you're still stuck never be afraid

52
00:03:53,600 --> 00:03:54,830
to ask a question.

53
00:03:54,920 --> 00:03:59,630
If you're sitting on something you're thinking haha I wish I knew the answer to this and wonder I can't

54
00:03:59,630 --> 00:04:01,320
find it anywhere.

55
00:04:01,430 --> 00:04:04,190
Just ask right if you're running into an issue like that.

56
00:04:04,340 --> 00:04:09,620
More than likely someone else is running into a similar issue so don't be afraid to ask pose the question

57
00:04:09,630 --> 00:04:14,330
you can go to stack overflow or discord or anywhere else can ask a question.

58
00:04:14,330 --> 00:04:17,320
Never be afraid to ask a question.

59
00:04:17,340 --> 00:04:18,100
All right.

60
00:04:18,230 --> 00:04:20,360
With that being said let's not wait any longer.

61
00:04:20,360 --> 00:04:23,860
Let's get our project folder set up and get into this second project.