They're Good Dogs A gentle introduction to Core ML and Vision - - PowerPoint PPT Presentation

▶
they re good dogs
SMART_READER_LITE
LIVE PREVIEW

They're Good Dogs A gentle introduction to Core ML and Vision - - PowerPoint PPT Presentation

Oct 31, 2023 430 likes •1.17k views

They're Good Dogs A gentle introduction to Core ML and Vision Andrew Harvey (@mootpointer) A Question For You What we will not cover In depth machine learning Cats Bleeding edge real - world implementations What we will cover

slide-1
SLIDE 1

They're Good Dogs

A gentle introduction to Core ML and Vision

Andrew Harvey (@mootpointer)

slide-2
SLIDE 2

A Question

For You

slide-3
SLIDE 3

What we will not cover

  • In depth machine learning
  • Cats
  • Bleeding edge real-world

implementations

slide-4
SLIDE 4

What we will cover

  • Dogs
  • The basics of the tools available
  • More dogs
  • What you could do with this stuff
slide-5
SLIDE 5
slide-6
SLIDE 6

Pro

Tip

slide-7
SLIDE 7

"The best part

  • f WWDC

is the labs."

slide-8
SLIDE 8

"The best part

  • f WWDC

is the labs !"

slide-9
SLIDE 9
slide-10
SLIDE 10
slide-11
SLIDE 11
slide-12
SLIDE 12
slide-13
SLIDE 13
slide-14
SLIDE 14
slide-15
SLIDE 15
slide-16
SLIDE 16
slide-17
SLIDE 17
slide-18
SLIDE 18
slide-19
SLIDE 19
slide-20
SLIDE 20
slide-21
SLIDE 21
slide-22
SLIDE 22
slide-23
SLIDE 23
slide-24
SLIDE 24

Enough dogs. Show me the code

(Said no one ever.)

slide-25
SLIDE 25

xkcd.com/1425

slide-26
SLIDE 26

Let's break it down

slide-27
SLIDE 27
slide-28
SLIDE 28

Your App

slide-29
SLIDE 29

Domain Specific

slide-30
SLIDE 30

Vision

slide-31
SLIDE 31

Vision

With a pixel buffer

VNImageRequestHandler(cvPixelBuffer: imageBuffer,

  • ptions: [:])
slide-32
SLIDE 32

Vision

Or a CGImage

VNImageRequestHandler(cgImage: image,

  • ptions: [:])
slide-33
SLIDE 33

Natural language processing

slide-34
SLIDE 34
slide-35
SLIDE 35

GameplayKit

slide-36
SLIDE 36

Core ML

slide-37
SLIDE 37

Core ML

The fun stuff!

model.prediction(from: featureProvider)

slide-38
SLIDE 38

MLFeatureProvider

slide-39
SLIDE 39

Metal Performance Shaders

slide-40
SLIDE 40

Accelerate and BNNs

slide-41
SLIDE 41

If you're starting out

You probably want

Core ML

slide-42
SLIDE 42

Any Questions?

slide-43
SLIDE 43

That's enough code.

More dogs!

slide-44
SLIDE 44

Let's talk about

Adversarial

Examples

slide-45
SLIDE 45
slide-46
SLIDE 46
slide-47
SLIDE 47
slide-48
SLIDE 48
slide-49
SLIDE 49
slide-50
SLIDE 50
slide-51
SLIDE 51

Here's one I prepared earlier

slide-52
SLIDE 52

(Live demo ensues)

slide-53
SLIDE 53

Ecosystem

slide-54
SLIDE 54

You still have to train your

  • wn model
slide-55
SLIDE 55

Keras

keras.io

slide-56
SLIDE 56

Caffe

caffe.berkeleyvision.org

slide-57
SLIDE 57

scikit-learn

scikit-learn.org

slide-58
SLIDE 58

coremltools

developer.apple.com/machine-learning

slide-59
SLIDE 59

So what?

slide-60
SLIDE 60

Privacy

slide-61
SLIDE 61

Latency

slide-62
SLIDE 62

Offline

slide-63
SLIDE 63

"If you can avoid a network round trip, you probably should."

slide-64
SLIDE 64

Core ML + ARKit = !

slide-65
SLIDE 65

Where to from here?

slide-66
SLIDE 66

Learn you some acronyms

CNNs, LSTMs, RNNs.

(Oh my!)

slide-67
SLIDE 67

xkcd.com/1838

slide-68
SLIDE 68

fast.ai

slide-69
SLIDE 69

Make your own.

developer.apple.com/machine-learning

slide-70
SLIDE 70

Understand your features

slide-71
SLIDE 71

Do

No Evil

slide-72
SLIDE 72

Make fun things!

slide-73
SLIDE 73

Thanks!

(I'm hiring!)

andrew@canceraid.com