1
00:00:00,360 --> 00:00:01,300
Welcome back.

2
00:00:01,350 --> 00:00:06,120
It's time to have a little bit of fun because you've just been listening to me talking for a bit and

3
00:00:06,180 --> 00:00:10,950
we want to get our hands dirty and try this machine learning ourselves.

4
00:00:10,950 --> 00:00:16,560
But even though we don't know much about it yet we can actually play with it.

5
00:00:16,560 --> 00:00:22,530
For example there's a great Web site here by Google called teachable machine.

6
00:00:22,740 --> 00:00:29,440
And as you can see what it does is well it helps you do some machine learning activities.

7
00:00:29,490 --> 00:00:34,620
So if I click on get started here and by the way by the time you watch the video maybe they've changed

8
00:00:34,620 --> 00:00:37,170
the layout of this Web site.

9
00:00:37,170 --> 00:00:43,740
It doesn't really matter because this is more for you to kind of understand how things work underneath

10
00:00:43,740 --> 00:00:46,440
the hood and even play with this yourself.

11
00:00:46,440 --> 00:00:55,530
So let's create an image project and for now all we're going to do is say hey I want to decide or create

12
00:00:55,530 --> 00:01:00,800
a machine learning model that detects faces or cats.

13
00:01:00,870 --> 00:01:06,450
So let's just say this is going to be faces and this is going to be cats.

14
00:01:06,840 --> 00:01:08,940
And we have to give it some image samples.

15
00:01:08,940 --> 00:01:10,770
So then the machine can learn.

16
00:01:10,860 --> 00:01:11,520
Right.

17
00:01:11,550 --> 00:01:15,920
So in here on my desktop I've actually saved a couple of photos.

18
00:01:15,930 --> 00:01:18,930
I found a nice little cute cat picture online.

19
00:01:19,500 --> 00:01:27,660
So let's upload the cat picture so well upload the cat picture here you can see that we're only uploading

20
00:01:27,660 --> 00:01:28,320
one.

21
00:01:28,560 --> 00:01:34,500
And here I will upload the faces picture while I upload the baby's picture here

22
00:01:38,760 --> 00:01:42,690
so we have faces and cats and then we train our model.

23
00:01:42,690 --> 00:01:47,280
We essentially tell the computer hey this is an example what a face looks like.

24
00:01:47,430 --> 00:01:49,790
And this is an example of what a cat looks like.

25
00:01:49,800 --> 00:01:51,920
So just learn please.

26
00:01:51,930 --> 00:01:55,760
So you can see over here it's learning.

27
00:01:55,830 --> 00:01:57,120
And there you go.

28
00:01:57,120 --> 00:01:59,830
It now has created a model.

29
00:02:00,090 --> 00:02:04,620
Now in here I can use my webcam or I can just upload a file.

30
00:02:04,620 --> 00:02:15,780
So let's just upload a snorkel picture that I have of myself and I click open.

31
00:02:15,790 --> 00:02:16,420
There you go.

32
00:02:16,420 --> 00:02:21,990
This model that we trained thinks that this has a forty nine percent chance.

33
00:02:22,030 --> 00:02:24,010
Let's make this a little bit bigger.

34
00:02:24,040 --> 00:02:32,080
This has a forty nine percent chance that this image is a face and a 51 percent chance that I'm a cat

35
00:02:34,100 --> 00:02:35,360
what's going on here.

36
00:02:35,360 --> 00:02:36,920
What if we give it a better picture.

37
00:02:36,920 --> 00:02:38,860
I mean this is a tough picture right.

38
00:02:38,860 --> 00:02:40,220
It's a little bit fuzzy.

39
00:02:40,460 --> 00:02:42,650
I have my snorkeling mask on.

40
00:02:42,740 --> 00:02:46,580
Maybe it's a tough picture for this model to predict.

41
00:02:46,580 --> 00:02:48,670
So I'm going to upload a different file.

42
00:02:48,680 --> 00:02:58,440
Let's say this exact same face that I use in the image if I click open Oh boy that's even worse.

43
00:02:58,440 --> 00:03:05,260
It thinks I'm 57 percent cat and it's 43 percent certain that this is a face.

44
00:03:05,310 --> 00:03:06,260
What's happening here.

45
00:03:07,470 --> 00:03:11,970
Well what you've just witnessed is how machine learning works.

46
00:03:11,970 --> 00:03:16,320
We provide the computer some data in this case.

47
00:03:16,320 --> 00:03:24,960
I provided one image of myself and Danielle's face and one picture of a cat based on those two inputs

48
00:03:25,080 --> 00:03:26,300
and data.

49
00:03:26,370 --> 00:03:33,240
I say Hey computer learn try to understand this picture and try to understand this picture.

50
00:03:33,420 --> 00:03:40,650
And next time I give you any images based on this data predict for me or tell me how much confidence

51
00:03:40,650 --> 00:03:48,310
you have that either it's a cat or a face and that's pretty much machine learning.

52
00:03:48,350 --> 00:03:50,700
It doesn't have to be just images.

53
00:03:50,750 --> 00:03:52,950
It just needs to be some sort of data.

54
00:03:53,000 --> 00:04:01,740
We trained the computer and we create a model that can predict for us based on that data.

55
00:04:01,760 --> 00:04:06,190
Now we see here that it's really really bad.

56
00:04:06,220 --> 00:04:09,390
I mean this is definitely not a cat right.

57
00:04:09,400 --> 00:04:12,970
So why is it bad.

58
00:04:13,090 --> 00:04:14,280
Think about this for a second.

59
00:04:14,290 --> 00:04:16,290
Try to answer it on your own.

60
00:04:16,360 --> 00:04:17,940
Ready for the answer.

61
00:04:17,950 --> 00:04:27,980
Well number one is that we've only given it one image each for a computer to really learn and to understand

62
00:04:27,980 --> 00:04:30,380
what a face is what a cat is.

63
00:04:30,440 --> 00:04:38,000
It needs hundreds thousands if not millions of images of cats to detect patterns to figure out hey this

64
00:04:38,000 --> 00:04:39,160
is what a cat is hate.

65
00:04:39,170 --> 00:04:41,410
This is what it faces right now.

66
00:04:41,420 --> 00:04:44,340
I've only given it one sample each.

67
00:04:44,390 --> 00:04:51,350
So this model is not very smart just like a human right to learn something in school you have to repeat

68
00:04:51,350 --> 00:04:51,490
it.

69
00:04:51,500 --> 00:04:53,580
You have to see it over and over.

70
00:04:53,780 --> 00:05:00,350
And then also with the preview sometimes I give it a really clear image of a face or sometimes I give

71
00:05:00,350 --> 00:05:08,500
it the snorkeling picture that's really fuzzy so that also affects the output but at the end of the

72
00:05:08,500 --> 00:05:15,640
day machine learning is just that we give it some input and then we tell the machine to learn based

73
00:05:15,640 --> 00:05:18,580
on those inputs and give us an output.

74
00:05:18,610 --> 00:05:22,350
Next time we'll give it a new input.

75
00:05:22,370 --> 00:05:24,290
Now I want you to take a break here.

76
00:05:24,290 --> 00:05:29,300
I'm going to link to this resource so that you can actually play with this yourself and see if I can

77
00:05:29,300 --> 00:05:30,680
improve this model.

78
00:05:30,680 --> 00:05:32,950
You can even have your web cam on.

79
00:05:32,950 --> 00:05:37,730
So instead of a file you have your web webcam just like in the examples that they provide.

80
00:05:37,730 --> 00:05:44,880
So if I go to teach a machine you can see here different projects that you can build just by clicking

81
00:05:44,880 --> 00:05:49,230
without programming anything you can even export your model.

82
00:05:49,260 --> 00:05:53,040
Once you're done so that you can use it in your own projects.

83
00:05:53,190 --> 00:05:58,680
Again if this is confusing don't worry we're actually going to go step by step and instead of just doing

84
00:05:58,680 --> 00:06:04,620
something visually we're actually going to code and learn how machine learning and data scientists work.

85
00:06:04,710 --> 00:06:10,930
But this is a great way for you to just understand on high level how machine learning works.

86
00:06:11,040 --> 00:06:15,620
So go off and play with this Web site and I'll see you in the next video by.