© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Build, train and deploy Deep
Learning models on Amazon
Sagemaker
Pedro Paez
Specialist Solutions Architect
AWS
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Related breakouts
Accelerating ML adoption with our new AI services
Ben Snively
Building Business outcome with Machine Learning
on AWS
Barnam Bora
ML in the physical world with IoT
Timothee Cruse
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Machine Learning
Y=f(X)
X=input / Y=output
Given many (X,Y) => Infer f
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Discriminating between classes
Positive
Negative
Cats
Dogs
x
x
x
x
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Making non-linear linear
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Biological learning
Source: http://cs231n.github.io/neural-networks-1/
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Deep learning
Hidden layers
Input layer Output
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
The “learning” in deep learning
0.4 0.3
0.2 0.9
...
...
backpropagation (gradient descent)
!" != !"
0.4 ± # 0.3 ± #
new
weights
new
weights
0
1
0
1
1
.
.
.
X
input
label
...
!"
...
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
General Concepts
• A neural network with at least one hidden layer can approximate
any function
• The optimization is complicated and computationally very intensive
due to non-convexity of the optimization space
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Trademark
M L F R A M E W O R K S &
I N F R A S T R U C T U R E
The Amazon ML Stack: Broadest & Deepest Set of Capabilities
A I S E R V I C E S
R E K O G N I T I O N
I M A G E
P O L L Y T R A N S C R I B E T R A N S L A T E C O M P R E H E N D
C O M P R E H E N D
M E D I C A L
L E XR E K O G N I T I O N
V I D E O
Vision Speech Chatbots
A M A Z O N S A G E M A K E R
B U I L D T R A I N
F O R E C A S TT E X T R A C T P E R S O N A L I Z E
D E P L O Y
Pre-built algorithms & notebooks
Data labeling (G R O U N D T R U T H )
One-click model training & tuning
Optimization ( N E O )
One-click deployment & hosting
M L S E R V I C E S
F r a m e w o r k s I n t e r f a c e s I n f r a s t r u c t u r e
E C 2 P 3
& P 3 d n
E C 2 C 5 F P G A s G R E E N G R A S S E L A S T I C
I N F E R E N C E
Models without training data (REINFORCEMENT LEARNING)Algorithms & models ( A W S M A R K E T P L A C E )
Language Forecasting Recommendations
NEW NEWNEW
NEW
NEW
NEWNEW
NEW
NEW
RL Coach
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Amazon SageMaker:
Build, Train, and Deploy ML Models at Scale
1
2
3
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Amazon SageMaker:
Build, Train, and Deploy ML Models at Scale
RL Coach
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Amazon SageMaker:
Build, Train, and Deploy ML Models at Scale
G R E E N G R A S SE C 2 C 5
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Amazon SageMaker:
Build, Train, and Deploy ML Models at Scale
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Pre-configured environments to quickly build deep learning
applications
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
AWS is framework agnostic
Choose from popular frameworks
Run them fully managed Or run them yourself
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
The best place to run TensorFlow
Fastest time
for TensorFlow
65% 90%
30m 14m
• 85% of TensorFlow workloads in the
cloud runs on AWS (2018 Nucleus
report)
• Available w/ Amazon SageMaker and
the AWS Deep Learning AMIs
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
AWS: Best platform to run PyTorch
Versatile PortableFlexible
© 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
AWS is the best platform for Apache MXNet
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
• Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial
Networks (UC Berkley):
• https://junyanz.github.io/CycleGAN/
https://github.com/dmlc/gluon-cv
https://github.com/pedrojpaez/summit_workshop
Thank you!
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Pedro Paez
pppaez@amazon.com

AWS Summit Singapore 2019 | Build, Train and Deploy Deep Learning Models on Amazon Sagemaker

  • 1.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Build, train and deploy Deep Learning models on Amazon Sagemaker Pedro Paez Specialist Solutions Architect AWS
  • 2.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Related breakouts Accelerating ML adoption with our new AI services Ben Snively Building Business outcome with Machine Learning on AWS Barnam Bora ML in the physical world with IoT Timothee Cruse
  • 3.
    S U MM I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 4.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Machine Learning Y=f(X) X=input / Y=output Given many (X,Y) => Infer f
  • 5.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Discriminating between classes Positive Negative Cats Dogs x x x x
  • 6.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Making non-linear linear
  • 7.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Biological learning Source: http://cs231n.github.io/neural-networks-1/
  • 8.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Deep learning Hidden layers Input layer Output
  • 9.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T The “learning” in deep learning 0.4 0.3 0.2 0.9 ... ... backpropagation (gradient descent) !" != !" 0.4 ± # 0.3 ± # new weights new weights 0 1 0 1 1 . . . X input label ... !" ...
  • 10.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T General Concepts • A neural network with at least one hidden layer can approximate any function • The optimization is complicated and computationally very intensive due to non-convexity of the optimization space
  • 11.
    S U MM I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 12.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon Trademark M L F R A M E W O R K S & I N F R A S T R U C T U R E The Amazon ML Stack: Broadest & Deepest Set of Capabilities A I S E R V I C E S R E K O G N I T I O N I M A G E P O L L Y T R A N S C R I B E T R A N S L A T E C O M P R E H E N D C O M P R E H E N D M E D I C A L L E XR E K O G N I T I O N V I D E O Vision Speech Chatbots A M A Z O N S A G E M A K E R B U I L D T R A I N F O R E C A S TT E X T R A C T P E R S O N A L I Z E D E P L O Y Pre-built algorithms & notebooks Data labeling (G R O U N D T R U T H ) One-click model training & tuning Optimization ( N E O ) One-click deployment & hosting M L S E R V I C E S F r a m e w o r k s I n t e r f a c e s I n f r a s t r u c t u r e E C 2 P 3 & P 3 d n E C 2 C 5 F P G A s G R E E N G R A S S E L A S T I C I N F E R E N C E Models without training data (REINFORCEMENT LEARNING)Algorithms & models ( A W S M A R K E T P L A C E ) Language Forecasting Recommendations NEW NEWNEW NEW NEW NEWNEW NEW NEW RL Coach
  • 13.
    S U MM I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 14.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Amazon SageMaker: Build, Train, and Deploy ML Models at Scale 1 2 3
  • 15.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Amazon SageMaker: Build, Train, and Deploy ML Models at Scale RL Coach
  • 16.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Amazon SageMaker: Build, Train, and Deploy ML Models at Scale G R E E N G R A S SE C 2 C 5
  • 17.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T Amazon SageMaker: Build, Train, and Deploy ML Models at Scale
  • 18.
    S U MM I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Pre-configured environments to quickly build deep learning applications
  • 19.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T AWS is framework agnostic Choose from popular frameworks Run them fully managed Or run them yourself
  • 20.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T The best place to run TensorFlow Fastest time for TensorFlow 65% 90% 30m 14m • 85% of TensorFlow workloads in the cloud runs on AWS (2018 Nucleus report) • Available w/ Amazon SageMaker and the AWS Deep Learning AMIs
  • 21.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T AWS: Best platform to run PyTorch Versatile PortableFlexible
  • 22.
    © 2019, AmazonWeb Services, Inc. or its affiliates. All rights reserved.S U M M I T AWS is the best platform for Apache MXNet
  • 23.
    S U MM I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 24.
    S U MM I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. • Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks (UC Berkley): • https://junyanz.github.io/CycleGAN/ https://github.com/dmlc/gluon-cv https://github.com/pedrojpaez/summit_workshop
  • 25.
    Thank you! S UM M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved. Pedro Paez pppaez@amazon.com