1

00:00:07,860 --> 00:00:10,010

Hey guys this is Caleb with Dev's Lopes dot com.



2

00:00:10,020 --> 00:00:13,680

And welcome to the Cornell section of the iOS 11 course.



3

00:00:13,740 --> 00:00:19,080

I'm so excited to get you into using Core M-L because it's seriously so cool.



4

00:00:19,080 --> 00:00:24,780

Apple has made it so easy to integrate machine learning into iOS apps and that's actually what you are



5

00:00:24,780 --> 00:00:25,530

here to learn.



6

00:00:25,530 --> 00:00:27,450

So let's dive right into it.



7

00:00:27,450 --> 00:00:33,690

I want to show you some parts of the apple developer documentation for core Armel and we're going to



8

00:00:33,690 --> 00:00:36,930

basically learn what you will learn in this section.



9

00:00:36,930 --> 00:00:41,100

I'm also going to tell you what Carmel is and why you should care as a developer.



10

00:00:41,160 --> 00:00:47,190

So I'm going to pull open Safari here and I'm just on the Carmel web site developer dot apple dot com



11

00:00:47,190 --> 00:00:49,300

slash machine learning if you're interested.



12

00:00:49,770 --> 00:00:55,410

And as you can see there are already a couple of ways that Apple is using machine learning.



13

00:00:55,410 --> 00:01:00,810

They use it in quick type in messages to predict what words you're going to type next.



14

00:01:00,810 --> 00:01:06,660

They use it in the camera to identify faces so that it can focus on your face automatically without



15

00:01:06,660 --> 00:01:08,630

you having to tap on the screen.



16

00:01:08,640 --> 00:01:13,890

They also use it here to identify landmarks that they use it in a lot of really cool ways already but



17

00:01:14,130 --> 00:01:16,050

you're here to learn it for yourself.



18

00:01:16,050 --> 00:01:20,420

So let's go ahead and let's just talk quickly about what is Corum out.



19

00:01:20,430 --> 00:01:26,280

Carmel is a massively helpful framework from Apple that allows you to put machine learning into your



20

00:01:26,280 --> 00:01:26,550

app.



21

00:01:26,550 --> 00:01:27,780

Simple as that.



22

00:01:27,840 --> 00:01:33,330

If we go here to the core AML page you can see it allows you to integrate machine learning models into



23

00:01:33,330 --> 00:01:37,560

your app and we're going to be using what's called a trained model.



24

00:01:37,760 --> 00:01:38,320

OK.



25

00:01:38,520 --> 00:01:43,760

And Apple provides some trained models that we can use in each of them kind of has a specific super



26

00:01:43,760 --> 00:01:44,630

power.



27

00:01:44,910 --> 00:01:48,090

One might be really good at identifying objects in a scene.



28

00:01:48,090 --> 00:01:53,190

One might be good at identifying faces and other one might be good at identifying like different types



29

00:01:53,190 --> 00:01:54,150

of scenes.



30

00:01:54,150 --> 00:01:57,180

OK we're going to basically use this train model.



31

00:01:57,180 --> 00:02:02,990

We're going to pass it an input case and maybe like a photo or a block of text it's going to run it



32

00:02:03,000 --> 00:02:08,670

through all kinds of complex calculations in the trained model and give us an output of our desired



33

00:02:08,670 --> 00:02:09,030

result.



34

00:02:09,030 --> 00:02:15,180

So if we took a photo of something it would pass the photo into the model and it would output a label



35

00:02:15,180 --> 00:02:19,200

or a piece of text with what it thinks that is in the photo.



36

00:02:19,200 --> 00:02:25,140

If we were to pass in a word that was a proper noun it could tell us that is someone's name it could



37

00:02:25,140 --> 00:02:28,680

identify that type of information.



38

00:02:28,680 --> 00:02:29,890

So that's really cool.



39

00:02:30,070 --> 00:02:31,310

So that's Corum out.



40

00:02:31,660 --> 00:02:36,520

There's another framework we're going to use called vision and this is really cool.



41

00:02:36,570 --> 00:02:43,020

It allows you to use what is called Computer Vision machine learning computer vision allows you to do



42

00:02:43,020 --> 00:02:46,470

things like face tracking face detection landmarks.



43

00:02:46,470 --> 00:02:50,020

Text detection rectangle detection barcode detection.



44

00:02:50,130 --> 00:02:54,110

Now as you're hearing this you're probably thinking of some apps you know that use this.



45

00:02:54,150 --> 00:02:59,250

If you've ever used an app called Word Lens I think that's what it was called before Google bought them



46

00:02:59,250 --> 00:03:03,990

and put it into Google translate but with Word Lens what you can do is you can take a picture of a street



47

00:03:03,990 --> 00:03:08,850

sign and it can translate that word in in real time on that street sign.



48

00:03:08,850 --> 00:03:09,930

It's very cool.



49

00:03:09,930 --> 00:03:13,840

So that kind of thing could be done with text detection with the vision framework.



50

00:03:13,920 --> 00:03:19,590

Not only that but it can do object tracking If you've ever posted on Instagram or snapchat story.



51

00:03:19,590 --> 00:03:25,680

You know that you can drop on an emoji or even words and you can pin them to a certain object in the



52

00:03:25,680 --> 00:03:26,300

frame.



53

00:03:26,400 --> 00:03:28,530

And then as you move that object follows you.



54

00:03:28,530 --> 00:03:31,180

That's how object tracking can be done with vision.



55

00:03:31,260 --> 00:03:36,600

And the last one is image registration and that's partly what we will use to create an app that lets



56

00:03:36,600 --> 00:03:41,580

us take a photo of something and then using that photo the app can identify what it is.



57

00:03:41,580 --> 00:03:43,200

It's really really cool.



58

00:03:43,560 --> 00:03:43,880

OK.



59

00:03:43,890 --> 00:03:44,830

So that's vision.



60

00:03:44,850 --> 00:03:49,560

And we've talked briefly about core M-L but the last one we're going to talk about is called and that's



61

00:03:49,650 --> 00:03:51,490

linguistic tagger.



62

00:03:51,590 --> 00:03:58,470

And that's a class that Apple has provided to allow us to analyze language and identify different things



63

00:03:58,530 --> 00:04:01,530

about it like we could identify proper names.



64

00:04:01,530 --> 00:04:09,670

We could do all kinds of other cooling gearstick stuff like we could determine whether a word is a noun



65

00:04:09,720 --> 00:04:14,730

if it's used a certain way in a sentence we can determine if it's a verb an adjective we can detect



66

00:04:14,730 --> 00:04:16,560

what language is being spoken.



67

00:04:16,560 --> 00:04:20,250

So imagine that if you're sitting somewhere you're not quite sure what language someone near you is



68

00:04:20,250 --> 00:04:24,900

speaking you can just hold up your phone and it could listen and detect Oh that's Spanish or that's



69

00:04:24,900 --> 00:04:25,460

Italian.



70

00:04:25,470 --> 00:04:27,600

K. There's lots of really cool ways.



71

00:04:27,600 --> 00:04:30,340

Machine learning can be used in your application.



72

00:04:30,390 --> 00:04:36,090

The coolest thing is that core M-L is stored locally on your device so it doesn't even need to pull



73

00:04:36,090 --> 00:04:40,300

data from some kind of cloud server or reach out to an API.



74

00:04:40,560 --> 00:04:43,950

The male model can do all of the thinking for us that's the coolest thing.



75

00:04:43,950 --> 00:04:49,620

It's sort of like we're putting a brain into our application and it's very very cool so I'm so excited



76

00:04:49,830 --> 00:04:50,870

to tell you more about it.



77

00:04:50,880 --> 00:04:55,500

But let's go ahead and let's move on to the next video so you can actually get started learning core



78

00:04:55,530 --> 00:04:56,200

m-L.



79

00:04:56,250 --> 00:04:57,860

All right so head over to the next video.



80

00:04:57,870 --> 00:04:58,470

We'll see you there.