1
00:00:00,330 --> 00:00:05,760
Hey, everybody, so we just saw the power of the Clarify API.

2
00:00:06,000 --> 00:00:07,930
It's pretty exciting, isn't it?

3
00:00:08,400 --> 00:00:15,210
Some really, really cool functionality and lots of possibilities now because I want to make sure that

4
00:00:15,210 --> 00:00:23,220
everybody gets through this course and you're able to successfully finish the app, deploy the app and

5
00:00:23,220 --> 00:00:24,750
show it off on your portfolio.

6
00:00:25,050 --> 00:00:29,970
I've made this video to just show you some of the common issues people might have.

7
00:00:29,970 --> 00:00:33,700
Also some of the power that you now have with clarify.

8
00:00:34,050 --> 00:00:35,510
So let's go through it.

9
00:00:36,240 --> 00:00:42,270
One is that I'm going to provide code for you at the end of each of the sections that are coming up

10
00:00:42,510 --> 00:00:48,780
as we build our Smart Brain app into more and more complex things with the back end, with the database,

11
00:00:49,320 --> 00:00:53,220
I'm going to provide code that you can always come back to.

12
00:00:53,370 --> 00:00:59,910
Let's say you are unable to code along or you don't know why your app isn't working.

13
00:01:00,750 --> 00:01:03,140
There's going to be different versions of the app.

14
00:01:03,570 --> 00:01:08,780
Obviously, this is going to be the first version that I'm going to link to at the end of this section.

15
00:01:09,480 --> 00:01:14,400
So one is the front end, which is called face recognition brain, at least for now.

16
00:01:14,670 --> 00:01:19,950
And then the next one is the face recognition brain API, which is the back end that we're going to

17
00:01:19,950 --> 00:01:21,870
build in later sections.

18
00:01:22,650 --> 00:01:25,130
But let me show you how things would work.

19
00:01:25,140 --> 00:01:32,790
Let's say we cloned both of these repos and they both have read me to show you how to get started again.

20
00:01:32,910 --> 00:01:38,340
Don't worry, because the API we're going to work on in later section and the front end, we still have

21
00:01:38,340 --> 00:01:39,180
a little bit to go.

22
00:01:39,850 --> 00:01:44,830
The one thing that you want to make sure is that you grab your own clarify API key.

23
00:01:45,630 --> 00:01:51,300
So clarify, although the website might change, is still going to have the same functionality.

24
00:01:51,450 --> 00:01:58,350
You're going to have to sign up if you don't have an account or you can just log in and once you log

25
00:01:58,350 --> 00:02:00,870
in, it's going to take you to their dashboard.

26
00:02:02,720 --> 00:02:07,670
Now, in my case, I already have an application built, but otherwise you can just click on create

27
00:02:07,970 --> 00:02:13,700
application and once you do that, it's going to ask you to put in the name some details.

28
00:02:13,940 --> 00:02:16,910
And then from there, you'll have this dashboard.

29
00:02:18,340 --> 00:02:25,420
Now, all we have to do is just click on Create New APIC, you can leave all the check boxes checked

30
00:02:25,540 --> 00:02:28,760
and just create an API key from here.

31
00:02:28,820 --> 00:02:30,840
This is your API key that you can use.

32
00:02:30,850 --> 00:02:31,930
So you just copy it.

33
00:02:33,710 --> 00:02:36,680
You added into the.

34
00:02:37,710 --> 00:02:39,570
Source app, dodgiest file.

35
00:02:42,030 --> 00:02:44,370
And you'll see here that I left it.

36
00:02:46,430 --> 00:02:49,260
For you to enter, it says your API key here.

37
00:02:49,310 --> 00:02:52,760
So you're going to replace that string with the API key.

38
00:03:01,750 --> 00:03:04,950
And you should be good to go, at least for now.

39
00:03:06,100 --> 00:03:12,850
Let's look at the map, so I'm just going to start the server and the front end.

40
00:03:13,240 --> 00:03:19,000
Don't worry, this part you haven't done yet, but I just want to debug the app with you just so you

41
00:03:19,000 --> 00:03:21,700
can see some of the common mistakes that people have.

42
00:03:28,180 --> 00:03:31,900
So I'm just going to sign in with an account that I've already created for this app.

43
00:03:32,940 --> 00:03:37,530
You see that I remember remembers my name, one thing is that Chrome has a new security feature.

44
00:03:38,520 --> 00:03:43,620
But again, this is our app that we're just starting off with, so you can just click, OK, if you

45
00:03:43,620 --> 00:03:45,140
get that prompt, not a big deal.

46
00:03:45,450 --> 00:03:48,480
And I have an image here, which is Tom Hanks.

47
00:03:49,080 --> 00:03:58,170
Let's copy that, making sure that we always copy a JPEG image, not a format that our app won't understand.

48
00:03:58,200 --> 00:04:02,640
So it's always a good idea to do a jpeg, an actual link to an image like this.

49
00:04:03,360 --> 00:04:06,900
And if I paste this in and I detect.

50
00:04:08,070 --> 00:04:13,470
Look at that, it detects the face and it puts a box around the face, very, very cool.

51
00:04:14,690 --> 00:04:22,220
Now, the cool thing with clarify is that, well, it's got multiple models that we can use in our case,

52
00:04:22,220 --> 00:04:24,220
we're using the face detection model.

53
00:04:24,680 --> 00:04:31,400
Now, when we use an API, sometimes a server goes down or an API doesn't work, especially since we're

54
00:04:31,400 --> 00:04:32,490
using the free version.

55
00:04:32,780 --> 00:04:37,160
So a good way to check if the API is working is to go to a link.

56
00:04:37,160 --> 00:04:44,690
I'll provide for you after this lecture where it actually checks the face detection model so you can

57
00:04:44,690 --> 00:04:47,690
actually see that the face detection model here is working for them.

58
00:04:47,990 --> 00:04:53,030
If it's not working well, then this blue box wouldn't appear on this website.

59
00:04:54,080 --> 00:04:57,160
So this is a good way to try and make sure that their model works.

60
00:04:58,490 --> 00:05:02,310
You can also go to their documentation, so this is what we're going to be using.

61
00:05:02,330 --> 00:05:09,170
We're going to be using the predict images and we go to the JavaScript tab to actually learn about the

62
00:05:09,170 --> 00:05:11,090
clarify models in the future.

63
00:05:11,100 --> 00:05:16,610
We can even do videos if we want, predicting whatever video information we feed it.

64
00:05:17,660 --> 00:05:23,100
And also this model gallery to show you that there's different things that we can do.

65
00:05:23,120 --> 00:05:24,470
We have celebrity model.

66
00:05:24,470 --> 00:05:25,970
We have apparel model.

67
00:05:26,720 --> 00:05:28,040
We have color model.

68
00:05:29,040 --> 00:05:31,710
So there's a lot of cool things that we can do.

69
00:05:33,520 --> 00:05:36,610
Such as the general model with our app.

70
00:05:37,670 --> 00:05:45,170
Now, the reason I'm showing you this is because sometimes a model might be working, might not be working,

71
00:05:45,170 --> 00:05:50,030
and I already shows you how to test if a model is working, but if you're getting any errors.

72
00:05:51,100 --> 00:05:58,450
I've provided a little rehme for you, especially when we call the clarify phase detect model, so another

73
00:05:58,450 --> 00:06:01,330
option is to use this piece of string.

74
00:06:01,360 --> 00:06:10,150
I've added comments here for you to decide so I could actually change this from clarify face to tech

75
00:06:10,150 --> 00:06:11,800
model to a string.

76
00:06:12,740 --> 00:06:18,430
And if the clarify face model isn't working, this should still work.

77
00:06:18,910 --> 00:06:20,560
So let me sign in again.

78
00:06:27,170 --> 00:06:28,310
Let's grab Tom Hanks.

79
00:06:29,310 --> 00:06:30,150
Face again.

80
00:06:31,530 --> 00:06:33,520
And you see that that is working as well.

81
00:06:34,050 --> 00:06:38,550
So that is another option in case the clarify model is a working.

82
00:06:40,170 --> 00:06:45,630
Or maybe you want to expand your app once you're done with the course and don't worry, you'll get there.

83
00:06:45,990 --> 00:06:49,300
You can try different models that they have.

84
00:06:49,770 --> 00:06:55,980
So remember, there's different types of face to tech models or different types of models that they

85
00:06:55,980 --> 00:06:56,200
have.

86
00:06:56,220 --> 00:07:00,210
So let's say we do the general model.

87
00:07:01,440 --> 00:07:07,830
And in here, let's just consider the response or say console log, hey, let's see what kind of response

88
00:07:07,830 --> 00:07:09,840
we get with a general model.

89
00:07:11,130 --> 00:07:14,220
So, again, I'm going to go back in to our app.

90
00:07:15,810 --> 00:07:17,940
Grab our good old friend, Tom Hanks.

91
00:07:20,510 --> 00:07:26,510
Pieces in here, I'm going to open up the developer console to see what we get and I had detect.

92
00:07:30,240 --> 00:07:34,560
All right, we don't have to worry about this error, it's the error is because we're not using the

93
00:07:34,560 --> 00:07:38,190
face detect model, so it doesn't understand how to put the box around it.

94
00:07:38,340 --> 00:07:41,140
But because we're using the general model.

95
00:07:41,160 --> 00:07:41,880
Check this out.

96
00:07:42,990 --> 00:07:50,730
If we now go to Output's Zero and data, this is the data that the general model returns for us.

97
00:07:50,760 --> 00:07:51,480
And check this out.

98
00:07:51,510 --> 00:07:52,440
This is really cool.

99
00:07:53,340 --> 00:08:00,360
It returns for us, concepts and concepts are look at that, it understands the picture or at least

100
00:08:00,360 --> 00:08:08,640
the model does it, saying the picture is of tag people, it's 90 percent confident of a politician.

101
00:08:08,670 --> 00:08:14,760
Ninety seven percent confidence of a portrait of a one man, of a business of administration.

102
00:08:14,880 --> 00:08:16,560
It's a press conference, banking.

103
00:08:17,070 --> 00:08:20,180
So kind of cool it understands the picture.

104
00:08:20,460 --> 00:08:28,030
So by just changing the model, I was able to learn new information about the picture.

105
00:08:28,140 --> 00:08:35,610
So after this course, if you want to explore and do something other than face, detect maybe an app

106
00:08:35,610 --> 00:08:40,440
that describes what the picture is and displays it, that's another cool option.

107
00:08:41,860 --> 00:08:47,980
So, again, the reason I am showing you this is that there's a lot of cool things that you can do with

108
00:08:47,980 --> 00:08:48,670
clarify.

109
00:08:48,790 --> 00:08:55,720
The reason I've used it for this course is because it allows all of you to build different apps that

110
00:08:55,840 --> 00:08:58,220
don't just look the same for everyone.

111
00:08:58,750 --> 00:09:01,230
So hopefully this gets you excited.

112
00:09:01,420 --> 00:09:05,410
And I'll link to a bunch of resources for you to check out that.

113
00:09:05,410 --> 00:09:10,330
Once you're done with the app, you're done with the course, you're going to have different ways to

114
00:09:10,330 --> 00:09:11,410
customize everything.

115
00:09:12,500 --> 00:09:14,150
All right, that's a lot of talk for me.

116
00:09:14,330 --> 00:09:15,310
I'll see you in the next one.

117
00:09:15,350 --> 00:09:15,760
Bye bye.