SlideShare a Scribd company logo
1 of 55
Download to read offline
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Guy Ernest
Principal Solution Arrchitect, AWS Deep Learning
Building Deep Learning Applications
with TensorFlow and SageMaker on
AWS
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Amazon business is built on deep
learning
Amazon is Hiring!
Roles: Machine Learning Scientists, Research Scientists,
Machine Learning Managers, Research Managers, Software
Engineers, Software Development Managers, Product
Managers, Technical Program Managers etc!
Locations: Sunnyvale, Palo Alto, Pasadena, Seattle,
Pittsburgh, Boston, New York City, Aachen Germany,
Cambridge UK, Turin Italy, Hyderabad, India
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
How many times did I say “data” so far?
Data Lakes from AWS
Analytics
• Unmatched durability, and availability at EB scale
• Best security, compliance, and audit capabilities
• Object-level controls for fine-grain access
• Fastest performance by retrieving subsets of data
• The most ways to bring data in
• 2x as many integrations with partners
• Analyze with broadest set of analytics & ML services
Machine
Learning
Real-time DataOn-premises
Data Lake
on AWS
MovementData Movement
Data Lakes, Analytics, and IoT Portfolio from
AWS
Broadest, deepest set of analytic services
Amazon SageMaker
AWS Deep Learning AMIs
Amazon Rekognition
Amazon Lex
AWS DeepLens
Amazon Comprehend
Amazon Translate
Amazon Transcribe
Amazon Polly
Amazon Athena
Amazon EMR
Amazon Redshift
Full-text search
Amazon Elasticsearch service
Amazon QuickSight
AWS Direct Connect
AWS Snowball
AWS Snowmobile
AWS Database Migration Service
AWS IoT Core
Amazon Kinesis Data Firehose
Amazon Kinesis Data Streams
Amazon Kinesis Video Streams
Data Lake
on AWS
Storage | Archival Storage | Data Catalog
AnalyticsMachine Learning
Real-time DataOn-premises MovementData Movement
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Agenda
• Advanced Topics in Amazon SageMaker
• Integration between Spark and Amazon SageMaker
• Amazon SageMaker Built-in Algorithm
– Time series forecasting using DeepAR Forecasting
– Image Classification (Transfer learning with ResNet)
• ML training and deployment using any ML framework (including TensorFlow)
• Hyper-parameters Optimization
• Q&A
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
What is Amazon SageMaker?
One-click training
for ML, DL, and
custom algorithms
Easier training with
hyperparameter
optimization
Highly-optimized
machine learning
algorithms
Deployment
without
engineering effort
Fully-managed
hosting at scale
Build
Pre-built
notebook
instances
Deploy
Train
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Amazon SageMaker
1 2 3 4
I I I I
Notebook Instances Algorithms ML Training Service ML Hosting Service
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Amazon SageMaker integration with
Apache Spark (on Amazon EMR)
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Calling Spark from Amazon SageMaker
https://aws.amazon.com/blogs/machine-learning/build-amazon-sagemaker-notebooks-backed-by-spark-in-amazon-emr/
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Calling Spark from Amazon SageMaker
https://aws.amazon.com/blogs/machine-learning/build-amazon-sagemaker-notebooks-backed-by-spark-in-amazon-emr/
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Calling Amazon SageMaker from Spark I
https://github.com/aws/sagemaker-spark
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Calling Amazon SageMaker from Spark II
https://github.com/aws/sagemaker-spark
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Integrating Amazon SageMaker in Spark
Pipeline
https://github.com/aws/sagemaker-spark
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Using Spark for Batch Prediction
https://aws.amazon.com/blogs/machine-learning/distributed-inference-using-apache-mxnet-and-apache-spark-on-amazon-emr/
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Using Spark for Batch Prediction
https://aws.amazon.com/blogs/machine-learning/distributed-inference-using-apache-mxnet-and-apache-spark-on-amazon-emr/
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
How to build ML algorithms to scale
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Scalable Architecture
GPU
GPU
GPU Local
State
Shared
State
Local
State
Local
State
File/Pipe Modes
Cloud Optimized
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Linear Learner
Regression:
Estimate a real valued function
Binary Classification:
Predict a 0/1 class
˜y = hw, xi + t ˜y =
(
1 if hw, xi > t
0 else
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Linear Learner
Regression (mean squared error)
SageMaker Other
1.02 1.06
1.09 1.02
0.332 0.183
0.086 0.129
83.3 84.5
Classification (F1 Score)
SageMaker Other
0.980 0.981
0.870 0.930
0.997 0.997
0.978 0.964
0.914 0.859
0.470 0.472
0.903 0.908
0.508 0.508
30 GB datasets for web-spam and web-url classification
0
0.2
0.4
0.6
0.8
1
1.2
0 5 10 15 20 25 30
CostinDollars
Billable time in Minutes
sagemaker-url sagemaker-spam other-url other-spam
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
K-Means Clustering
Method Accurate? Passes Efficient
Tuning
Comments
Lloyds [1] Yes* 5-10 No
K-Means ++ [2] Yes k+5 to k+10 No scikit-learn
K-Means|| [3] Yes 7-12 No spark.ml
Online [4] No 1 No
Streaming [5,6] No 1 No Impractical
Webscale [7] No 1 No spark streaming
Coresets [8] No 1 Yes Impractical
SageMaker Yes 1 Yes
[1] Lloyd, IEEE TIT, 1982
[2] Arthur et. al. ACM-SIAM, 2007
[3] Bahmani et. al., VLDB, 2012
[4] Liberty et. al., 2015
[5] Shindler et. al, NIPS, 2011
[6] Guha et. al, IEEE Trans. Knowl. Data Eng. 2003
[7] Sculley, WWW, 2010
[8] Feldman et. al.
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
0
1
2
3
4
5
6
7
8
10 100 500
BillableTimeinMinutes Number of Clusters
sagemaker other
K-Means Clustering
k SageMaker Other
Text
1.2GB
10 1.18E3 1.18E3
100 1.00E3 9.77E2
500 9.18.E2 9.03E2
Images
9GB
10 3.29E2 3.28E2
100 2.72E2 2.71E2
500 2.17E2 Failed
Videos
27GB
10 2.19E2 2.18E2
100 2.03E2 2.02E2
500 1.86E2 1.85E2
Advertising
127GB
10 1.72E7 Failed
100 1.30E7 Failed
500 1.03E7 Failed
Synthetic
1100GB
10 3.81E7 Failed
100 3.51E7 Failed
500 2.81E7 Failed
Running Time vs. Number of Clusters
~10x Faster!
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithm Example
amazon-sagemaker-examples/introduction_to_amazon_algorithms/deepar_synthetic
Time Series Forecasting using DeepAR Forecasting algorithm
Amazon uses forecasting to drive business decisions
• Forecasting the product and labor in fulfillment centers
• Elastically scaling AWS compute and storage capacity for AWS customers
Benefits
• Cold start forecasting – forecasting on a new time series using multiple related
time series in training data
• Probabilistic forecasting – forecasting “the amount of sneakers sold in a week is
between X and Y with Z% probability”
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms Example
amazon-sagemaker-examples/introduction_to_amazon_algorithms/deepar_synthetic
Time Series Forecasting using DeepAR Forecasting algorithm
Training Data
Test Data
Training at
SageMaker
Model Artifact
Deployment
at SageMaker
Amazon SageMakerAmazon ECR
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms
amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb
Classifying images using Image Classification algorithm
Training Mode
• Full training using your own data only
• Transfer learning using your own data with ImageNet pre-trained model
Demo Scenario
• CALTECH 256 Training dataset (30607 images with 256 categories)
(http://www.vision.caltech.edu/Image_Datasets/Caltech256/)
• Use 15420 images for training
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms – Create Job
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms – Hyper-Parameters
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms – Input Data
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms – Input Data II
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms (2/2)
label - mars
probability – 0. 1373346447944641
Inference result of a model trained from scratch
amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb
Classifying images using Image Classification algorithm
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms (2/2)
label - mars
probability – 0. 1373346447944641
Inference result of a model trained from scratch
amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb
Classifying images using Image Classification algorithm
15,420 training images are NOT sufficient.
Gather more training
images
Conduct Transfer
Learning
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms (2/2)
amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb
Classifying images using Image Classification algorithm
Conduct Transfer
Learning
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Built-In ML Algorithms (2/2)
amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb
Classifying images using Image Classification algorithm
label – bathtub
probability - 0.998812198638916
Inference result of a model from Transfer Learning
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Bring Your Own Script
(TensorFlow/Apache MXNet)
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Bring Your Own Script (TensorFlow)
amazon-sagemaker-workshop/TensorFlow_Distributed_MNIST.ipynb
Deep Learning Framework: TensorFlow
Neural Network Model: Bring your own network
Dataset: MNIST
Training Mode: Distributed Training with 2 or more nodes
model_fn
MNIST
Training Data
mnist.py
Defines network architecture
Main script used for training and hosting
• model_fn : a function that contains all the logic to support training, evaluation, and
prediction
• train_input_fn : preprocess and load training data
• eval_input_fn : preprocess and load evaluation data
• serving_input_fn : defines the features to be passed to the model during prediction
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Bring Your Own Script (TensorFlow)
amazon-sagemaker-examples/sagemaker-python-sdk/tensorflow_resnet_cifar10_with_tensorboard
Deep Learning Framework: TensorFlow
Neural Network Model: Bring your own model
Dataset: MNIST
Training Mode: Distributed Training with 2 or more nodes
MNIST
Training Data
Training at
SageMaker
Model Artifact
Deployment
at SageMaker
Amazon SageMaker
Mnist.py
Amazon ECR
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Bring Your Own Container (Any ML/DL
Framework)
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Bring Your Own Container (R)
sample-notebooks/advanced_functionality/r_bring_your_own
Any Machine Learning or Deep Learning framework
Any Programming Language
mars.R
plumber.R
Dockerfile
Functions for training (fit) or serving the model
Using plumber package to create a lightweight HTTP server for serving
Defining a Docker container image for Training and Hosting
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Bring Your Own Container (R)
sample-notebooks/advanced_functionality/r_bring_your_own
Use the same method for training and hosting model
mars.R
plumber.R
Dockerfile
Build&PushDocker
Image
Amazon ECR
Training Data
Training at
SageMaker
Model Artifact
Deployment
at SageMaker
Amazon SageMaker
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Hyper-Parameter Optimization (HPO)
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Tuning Job Parameters
{
"Strategy": "Line"|"Random"|"Fill”|"Bayesian",
"TuningJobObjective": {
"Type": "Maximize"|"Minimize",
"MetricName": "string"
},
"ResourceLimits": {
"MaxNumberOfTrainingJobs": integer,
"MaxParallelTrainingJobs": integer,
"MaxClockTimeInHours": integer },
"ParameterRanges": {...}
}
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Grid Search vs. Random Search
Core illustration from Random Search for Hyper-Parameter Optimization by Bergstra and Bengio.
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Random Search vs. Fill Search
© 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
Bayesian Search
Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Tel Aviv Summit 2018

More Related Content

What's hot

Integrating Amazon SageMaker into your Enterprise - AWS Online Tech Talks
Integrating Amazon SageMaker into your Enterprise - AWS Online Tech TalksIntegrating Amazon SageMaker into your Enterprise - AWS Online Tech Talks
Integrating Amazon SageMaker into your Enterprise - AWS Online Tech TalksAmazon Web Services
 
Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...
Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...
Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...Amazon Web Services
 
Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...
Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...
Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...Amazon Web Services
 
Supercharge your Machine Learning Solutions with Amazon SageMaker
Supercharge your Machine Learning Solutions with Amazon SageMakerSupercharge your Machine Learning Solutions with Amazon SageMaker
Supercharge your Machine Learning Solutions with Amazon SageMakerAmazon Web Services
 
Introduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelIntroduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelAmazon Web Services
 
AWS Startup Day Kyiv - AI/ML services for developers
AWS Startup Day Kyiv - AI/ML services for developersAWS Startup Day Kyiv - AI/ML services for developers
AWS Startup Day Kyiv - AI/ML services for developersAmazon Web Services
 
Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...
Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...
Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...Amazon Web Services
 
Amazon SageMaker Build, Train and Deploy Your ML Models
Amazon SageMaker Build, Train and Deploy Your ML ModelsAmazon SageMaker Build, Train and Deploy Your ML Models
Amazon SageMaker Build, Train and Deploy Your ML ModelsAWS Riyadh User Group
 
Building a Modern Data Platform in the Cloud
Building a Modern Data Platform in the CloudBuilding a Modern Data Platform in the Cloud
Building a Modern Data Platform in the CloudAmazon Web Services
 
Machine Learning for innovation and transformation
Machine Learning for innovation and transformationMachine Learning for innovation and transformation
Machine Learning for innovation and transformationAmazon Web Services
 
An Overview of Machine Learning on AWS
An Overview of Machine Learning on AWSAn Overview of Machine Learning on AWS
An Overview of Machine Learning on AWSAmazon Web Services
 
Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...
Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...
Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...Amazon Web Services
 
Tensors for topic modeling and deep learning on AWS Sagemaker
Tensors for topic modeling and deep learning on AWS SagemakerTensors for topic modeling and deep learning on AWS Sagemaker
Tensors for topic modeling and deep learning on AWS SagemakerAnima Anandkumar
 
Work with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS SummitWork with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS SummitAmazon Web Services
 
[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...
[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...
[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...Amazon Web Services
 

What's hot (20)

Integrating Amazon SageMaker into your Enterprise - AWS Online Tech Talks
Integrating Amazon SageMaker into your Enterprise - AWS Online Tech TalksIntegrating Amazon SageMaker into your Enterprise - AWS Online Tech Talks
Integrating Amazon SageMaker into your Enterprise - AWS Online Tech Talks
 
Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...
Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...
Create an ML Factory in Financial Services with CI CD - FSI301 - New York AWS...
 
Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...
Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...
Using Amazon SageMaker and AWS DeepLens with Teams New to Machine Learning (G...
 
Supercharge your Machine Learning Solutions with Amazon SageMaker
Supercharge your Machine Learning Solutions with Amazon SageMakerSupercharge your Machine Learning Solutions with Amazon SageMaker
Supercharge your Machine Learning Solutions with Amazon SageMaker
 
Introduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day IsraelIntroduction to AI services for Developers - Builders Day Israel
Introduction to AI services for Developers - Builders Day Israel
 
AWS Startup Day Kyiv - AI/ML services for developers
AWS Startup Day Kyiv - AI/ML services for developersAWS Startup Day Kyiv - AI/ML services for developers
AWS Startup Day Kyiv - AI/ML services for developers
 
Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...
Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...
Intelligence of Things: IoT, AWS DeepLens and Amazon SageMaker - AWS Summit S...
 
Amazon SageMaker Build, Train and Deploy Your ML Models
Amazon SageMaker Build, Train and Deploy Your ML ModelsAmazon SageMaker Build, Train and Deploy Your ML Models
Amazon SageMaker Build, Train and Deploy Your ML Models
 
Building a Modern Data Platform in the Cloud
Building a Modern Data Platform in the CloudBuilding a Modern Data Platform in the Cloud
Building a Modern Data Platform in the Cloud
 
Amazon SageMaker In Action
Amazon SageMaker In Action Amazon SageMaker In Action
Amazon SageMaker In Action
 
Machine Learning for innovation and transformation
Machine Learning for innovation and transformationMachine Learning for innovation and transformation
Machine Learning for innovation and transformation
 
An Overview of Machine Learning on AWS
An Overview of Machine Learning on AWSAn Overview of Machine Learning on AWS
An Overview of Machine Learning on AWS
 
Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...
Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...
Market Prediction Using ML: Experiment with Amazon SageMaker and the Deutsche...
 
Migrating database to cloud
Migrating database to cloudMigrating database to cloud
Migrating database to cloud
 
Tensors for topic modeling and deep learning on AWS Sagemaker
Tensors for topic modeling and deep learning on AWS SagemakerTensors for topic modeling and deep learning on AWS Sagemaker
Tensors for topic modeling and deep learning on AWS Sagemaker
 
Work with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS SummitWork with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Toronto AWS Summit
 
[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...
[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...
[NEW LAUNCH!] Introducing Amazon SageMaker RL - Build and Train Reinforcement...
 
Intro to SageMaker
Intro to SageMakerIntro to SageMaker
Intro to SageMaker
 
VMware cloud on AWS
VMware cloud on AWSVMware cloud on AWS
VMware cloud on AWS
 
AWS reInvent 2018 recap edition
AWS reInvent 2018 recap editionAWS reInvent 2018 recap edition
AWS reInvent 2018 recap edition
 

Similar to Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Tel Aviv Summit 2018

Work with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS SummitWork with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS SummitAmazon Web Services
 
AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...
AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...
AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...Amazon Web Services Korea
 
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...Amazon Web Services
 
Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...
Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...
Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...Amazon Web Services
 
Speed up your Machine Learning workflows with build-in algorithms
Speed up your Machine Learning workflows with build-in algorithmsSpeed up your Machine Learning workflows with build-in algorithms
Speed up your Machine Learning workflows with build-in algorithmsJulien SIMON
 
CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...
CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...
CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...Amazon Web Services
 
Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências - MCL...
Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências -  MCL...Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências -  MCL...
Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências - MCL...Amazon Web Services
 
Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...
Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...
Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...Amazon Web Services
 
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018Amazon Web Services
 
Quickly and easily build, train, and deploy machine learning models at any scale
Quickly and easily build, train, and deploy machine learning models at any scaleQuickly and easily build, train, and deploy machine learning models at any scale
Quickly and easily build, train, and deploy machine learning models at any scaleAWS Germany
 
AWS Machine Learning Week SF: End to End Model Development Using SageMaker
AWS Machine Learning Week SF: End to End Model Development Using SageMakerAWS Machine Learning Week SF: End to End Model Development Using SageMaker
AWS Machine Learning Week SF: End to End Model Development Using SageMakerAmazon Web Services
 
Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...
Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...
Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...Amazon Web Services
 
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...Amazon Web Services
 
BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...
BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...
BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...Amazon Web Services
 
AWS re:Invent 2018 - ENT321 - SageMaker Workshop
AWS re:Invent 2018 - ENT321 - SageMaker WorkshopAWS re:Invent 2018 - ENT321 - SageMaker Workshop
AWS re:Invent 2018 - ENT321 - SageMaker WorkshopJulien SIMON
 
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...Amazon Web Services
 
Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...
Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...
Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...Amazon Web Services
 
From Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMakerFrom Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMakerAmazon Web Services
 
Supercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMakerSupercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMakerAmazon Web Services
 
End to End Model Development to Deployment using SageMaker
End to End Model Development to Deployment using SageMakerEnd to End Model Development to Deployment using SageMaker
End to End Model Development to Deployment using SageMakerAmazon Web Services
 

Similar to Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Tel Aviv Summit 2018 (20)

Work with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS SummitWork with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS Summit
Work with Machine Learning in Amazon SageMaker - BDA203 - Atlanta AWS Summit
 
AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...
AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...
AWS의 새로운 언어, 음성, 텍스트 처리 인공 지능 서비스, Amazon SageMaker::Sunil Mallya::AWS Summit...
 
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
Get Started with Deep Learning and Computer Vision Using AWS DeepLens (AIM316...
 
Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...
Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...
Speed up your Machine Learning workflows with built-in algorithms - Tel Aviv ...
 
Speed up your Machine Learning workflows with build-in algorithms
Speed up your Machine Learning workflows with build-in algorithmsSpeed up your Machine Learning workflows with build-in algorithms
Speed up your Machine Learning workflows with build-in algorithms
 
CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...
CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...
CI/CD for Your Machine Learning Pipeline with Amazon SageMaker (DVC303) - AWS...
 
Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências - MCL...
Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências -  MCL...Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências -  MCL...
Machine Learning e Amazon SageMaker: Algoritmos, Modelos e Inferências - MCL...
 
Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...
Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...
Perform Machine Learning at the IoT Edge using AWS Greengrass and Amazon Sage...
 
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
Demystifying Machine Learning On AWS - AWS Summit Sydney 2018
 
Quickly and easily build, train, and deploy machine learning models at any scale
Quickly and easily build, train, and deploy machine learning models at any scaleQuickly and easily build, train, and deploy machine learning models at any scale
Quickly and easily build, train, and deploy machine learning models at any scale
 
AWS Machine Learning Week SF: End to End Model Development Using SageMaker
AWS Machine Learning Week SF: End to End Model Development Using SageMakerAWS Machine Learning Week SF: End to End Model Development Using SageMaker
AWS Machine Learning Week SF: End to End Model Development Using SageMaker
 
Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...
Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...
Building, Training, and Deploying fast.ai Models Using Amazon SageMaker (AIM4...
 
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
Build Deep Learning Applications Using Apache MXNet - Featuring Chick-fil-A (...
 
BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...
BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...
BDA301 Working with Machine Learning in Amazon SageMaker: Algorithms, Models,...
 
AWS re:Invent 2018 - ENT321 - SageMaker Workshop
AWS re:Invent 2018 - ENT321 - SageMaker WorkshopAWS re:Invent 2018 - ENT321 - SageMaker Workshop
AWS re:Invent 2018 - ENT321 - SageMaker Workshop
 
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
Build Deep Learning Applications Using Apache MXNet, Featuring Workday (AIM40...
 
Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...
Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...
Accelerate ML Training on Amazon SageMaker Using GPU-Based EC2 P3 Instances (...
 
From Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMakerFrom Notebook to production with Amazon SageMaker
From Notebook to production with Amazon SageMaker
 
Supercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMakerSupercharge Your Machine Learning Model with Amazon SageMaker
Supercharge Your Machine Learning Model with Amazon SageMaker
 
End to End Model Development to Deployment using SageMaker
End to End Model Development to Deployment using SageMakerEnd to End Model Development to Deployment using SageMaker
End to End Model Development to Deployment using SageMaker
 

More from Amazon Web Services

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Amazon Web Services
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Amazon Web Services
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateAmazon Web Services
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSAmazon Web Services
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Amazon Web Services
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Amazon Web Services
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...Amazon Web Services
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsAmazon Web Services
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareAmazon Web Services
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSAmazon Web Services
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAmazon Web Services
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareAmazon Web Services
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWSAmazon Web Services
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckAmazon Web Services
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without serversAmazon Web Services
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...Amazon Web Services
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceAmazon Web Services
 

More from Amazon Web Services (20)

Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
Come costruire servizi di Forecasting sfruttando algoritmi di ML e deep learn...
 
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
Big Data per le Startup: come creare applicazioni Big Data in modalità Server...
 
Esegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS FargateEsegui pod serverless con Amazon EKS e AWS Fargate
Esegui pod serverless con Amazon EKS e AWS Fargate
 
Costruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWSCostruire Applicazioni Moderne con AWS
Costruire Applicazioni Moderne con AWS
 
Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot Come spendere fino al 90% in meno con i container e le istanze spot
Come spendere fino al 90% in meno con i container e le istanze spot
 
Open banking as a service
Open banking as a serviceOpen banking as a service
Open banking as a service
 
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
Rendi unica l’offerta della tua startup sul mercato con i servizi Machine Lea...
 
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...OpsWorks Configuration Management: automatizza la gestione e i deployment del...
OpsWorks Configuration Management: automatizza la gestione e i deployment del...
 
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows WorkloadsMicrosoft Active Directory su AWS per supportare i tuoi Windows Workloads
Microsoft Active Directory su AWS per supportare i tuoi Windows Workloads
 
Computer Vision con AWS
Computer Vision con AWSComputer Vision con AWS
Computer Vision con AWS
 
Database Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatareDatabase Oracle e VMware Cloud on AWS i miti da sfatare
Database Oracle e VMware Cloud on AWS i miti da sfatare
 
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJSCrea la tua prima serverless ledger-based app con QLDB e NodeJS
Crea la tua prima serverless ledger-based app con QLDB e NodeJS
 
API moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e webAPI moderne real-time per applicazioni mobili e web
API moderne real-time per applicazioni mobili e web
 
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatareDatabase Oracle e VMware Cloud™ on AWS: i miti da sfatare
Database Oracle e VMware Cloud™ on AWS: i miti da sfatare
 
Tools for building your MVP on AWS
Tools for building your MVP on AWSTools for building your MVP on AWS
Tools for building your MVP on AWS
 
How to Build a Winning Pitch Deck
How to Build a Winning Pitch DeckHow to Build a Winning Pitch Deck
How to Build a Winning Pitch Deck
 
Building a web application without servers
Building a web application without serversBuilding a web application without servers
Building a web application without servers
 
Fundraising Essentials
Fundraising EssentialsFundraising Essentials
Fundraising Essentials
 
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
AWS_HK_StartupDay_Building Interactive websites while automating for efficien...
 
Introduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container ServiceIntroduzione a Amazon Elastic Container Service
Introduzione a Amazon Elastic Container Service
 

Building Deep Learning Applications with TensorFlow and SageMaker on AWS - Tel Aviv Summit 2018

  • 1. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. Guy Ernest Principal Solution Arrchitect, AWS Deep Learning Building Deep Learning Applications with TensorFlow and SageMaker on AWS
  • 2. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. Amazon business is built on deep learning
  • 3.
  • 4.
  • 5.
  • 6.
  • 7. Amazon is Hiring! Roles: Machine Learning Scientists, Research Scientists, Machine Learning Managers, Research Managers, Software Engineers, Software Development Managers, Product Managers, Technical Program Managers etc! Locations: Sunnyvale, Palo Alto, Pasadena, Seattle, Pittsburgh, Boston, New York City, Aachen Germany, Cambridge UK, Turin Italy, Hyderabad, India
  • 8. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. How many times did I say “data” so far?
  • 9. Data Lakes from AWS Analytics • Unmatched durability, and availability at EB scale • Best security, compliance, and audit capabilities • Object-level controls for fine-grain access • Fastest performance by retrieving subsets of data • The most ways to bring data in • 2x as many integrations with partners • Analyze with broadest set of analytics & ML services Machine Learning Real-time DataOn-premises Data Lake on AWS MovementData Movement
  • 10. Data Lakes, Analytics, and IoT Portfolio from AWS Broadest, deepest set of analytic services Amazon SageMaker AWS Deep Learning AMIs Amazon Rekognition Amazon Lex AWS DeepLens Amazon Comprehend Amazon Translate Amazon Transcribe Amazon Polly Amazon Athena Amazon EMR Amazon Redshift Full-text search Amazon Elasticsearch service Amazon QuickSight AWS Direct Connect AWS Snowball AWS Snowmobile AWS Database Migration Service AWS IoT Core Amazon Kinesis Data Firehose Amazon Kinesis Data Streams Amazon Kinesis Video Streams Data Lake on AWS Storage | Archival Storage | Data Catalog AnalyticsMachine Learning Real-time DataOn-premises MovementData Movement
  • 11. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Agenda • Advanced Topics in Amazon SageMaker • Integration between Spark and Amazon SageMaker • Amazon SageMaker Built-in Algorithm – Time series forecasting using DeepAR Forecasting – Image Classification (Transfer learning with ResNet) • ML training and deployment using any ML framework (including TensorFlow) • Hyper-parameters Optimization • Q&A
  • 12. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. What is Amazon SageMaker? One-click training for ML, DL, and custom algorithms Easier training with hyperparameter optimization Highly-optimized machine learning algorithms Deployment without engineering effort Fully-managed hosting at scale Build Pre-built notebook instances Deploy Train
  • 13. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Amazon SageMaker 1 2 3 4 I I I I Notebook Instances Algorithms ML Training Service ML Hosting Service
  • 14. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. Amazon SageMaker integration with Apache Spark (on Amazon EMR)
  • 15. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Calling Spark from Amazon SageMaker https://aws.amazon.com/blogs/machine-learning/build-amazon-sagemaker-notebooks-backed-by-spark-in-amazon-emr/
  • 16. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Calling Spark from Amazon SageMaker https://aws.amazon.com/blogs/machine-learning/build-amazon-sagemaker-notebooks-backed-by-spark-in-amazon-emr/
  • 17. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Calling Amazon SageMaker from Spark I https://github.com/aws/sagemaker-spark
  • 18. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Calling Amazon SageMaker from Spark II https://github.com/aws/sagemaker-spark
  • 19. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Integrating Amazon SageMaker in Spark Pipeline https://github.com/aws/sagemaker-spark
  • 20. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Using Spark for Batch Prediction https://aws.amazon.com/blogs/machine-learning/distributed-inference-using-apache-mxnet-and-apache-spark-on-amazon-emr/
  • 21. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Using Spark for Batch Prediction https://aws.amazon.com/blogs/machine-learning/distributed-inference-using-apache-mxnet-and-apache-spark-on-amazon-emr/
  • 22. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. How to build ML algorithms to scale
  • 23. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 24. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Scalable Architecture GPU GPU GPU Local State Shared State Local State Local State File/Pipe Modes Cloud Optimized
  • 25. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Linear Learner Regression: Estimate a real valued function Binary Classification: Predict a 0/1 class ˜y = hw, xi + t ˜y = ( 1 if hw, xi > t 0 else
  • 26. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Linear Learner Regression (mean squared error) SageMaker Other 1.02 1.06 1.09 1.02 0.332 0.183 0.086 0.129 83.3 84.5 Classification (F1 Score) SageMaker Other 0.980 0.981 0.870 0.930 0.997 0.997 0.978 0.964 0.914 0.859 0.470 0.472 0.903 0.908 0.508 0.508 30 GB datasets for web-spam and web-url classification 0 0.2 0.4 0.6 0.8 1 1.2 0 5 10 15 20 25 30 CostinDollars Billable time in Minutes sagemaker-url sagemaker-spam other-url other-spam
  • 27. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. K-Means Clustering Method Accurate? Passes Efficient Tuning Comments Lloyds [1] Yes* 5-10 No K-Means ++ [2] Yes k+5 to k+10 No scikit-learn K-Means|| [3] Yes 7-12 No spark.ml Online [4] No 1 No Streaming [5,6] No 1 No Impractical Webscale [7] No 1 No spark streaming Coresets [8] No 1 Yes Impractical SageMaker Yes 1 Yes [1] Lloyd, IEEE TIT, 1982 [2] Arthur et. al. ACM-SIAM, 2007 [3] Bahmani et. al., VLDB, 2012 [4] Liberty et. al., 2015 [5] Shindler et. al, NIPS, 2011 [6] Guha et. al, IEEE Trans. Knowl. Data Eng. 2003 [7] Sculley, WWW, 2010 [8] Feldman et. al.
  • 28. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. 0 1 2 3 4 5 6 7 8 10 100 500 BillableTimeinMinutes Number of Clusters sagemaker other K-Means Clustering k SageMaker Other Text 1.2GB 10 1.18E3 1.18E3 100 1.00E3 9.77E2 500 9.18.E2 9.03E2 Images 9GB 10 3.29E2 3.28E2 100 2.72E2 2.71E2 500 2.17E2 Failed Videos 27GB 10 2.19E2 2.18E2 100 2.03E2 2.02E2 500 1.86E2 1.85E2 Advertising 127GB 10 1.72E7 Failed 100 1.30E7 Failed 500 1.03E7 Failed Synthetic 1100GB 10 3.81E7 Failed 100 3.51E7 Failed 500 2.81E7 Failed Running Time vs. Number of Clusters ~10x Faster!
  • 29. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 30. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithm Example amazon-sagemaker-examples/introduction_to_amazon_algorithms/deepar_synthetic Time Series Forecasting using DeepAR Forecasting algorithm Amazon uses forecasting to drive business decisions • Forecasting the product and labor in fulfillment centers • Elastically scaling AWS compute and storage capacity for AWS customers Benefits • Cold start forecasting – forecasting on a new time series using multiple related time series in training data • Probabilistic forecasting – forecasting “the amount of sneakers sold in a week is between X and Y with Z% probability”
  • 31. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms Example amazon-sagemaker-examples/introduction_to_amazon_algorithms/deepar_synthetic Time Series Forecasting using DeepAR Forecasting algorithm Training Data Test Data Training at SageMaker Model Artifact Deployment at SageMaker Amazon SageMakerAmazon ECR
  • 32. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb Classifying images using Image Classification algorithm Training Mode • Full training using your own data only • Transfer learning using your own data with ImageNet pre-trained model Demo Scenario • CALTECH 256 Training dataset (30607 images with 256 categories) (http://www.vision.caltech.edu/Image_Datasets/Caltech256/) • Use 15420 images for training
  • 33. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms – Create Job
  • 34. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms – Hyper-Parameters
  • 35. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms – Input Data
  • 36. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms – Input Data II
  • 37. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms (2/2) label - mars probability – 0. 1373346447944641 Inference result of a model trained from scratch amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb Classifying images using Image Classification algorithm
  • 38. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms (2/2) label - mars probability – 0. 1373346447944641 Inference result of a model trained from scratch amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb Classifying images using Image Classification algorithm 15,420 training images are NOT sufficient. Gather more training images Conduct Transfer Learning
  • 39.
  • 40. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms (2/2) amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb Classifying images using Image Classification algorithm Conduct Transfer Learning
  • 41. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Built-In ML Algorithms (2/2) amazon-sagemaker-workshop/Image-classification-transfer-learning.ipynb Classifying images using Image Classification algorithm label – bathtub probability - 0.998812198638916 Inference result of a model from Transfer Learning
  • 42.
  • 43. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. Bring Your Own Script (TensorFlow/Apache MXNet)
  • 44. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Bring Your Own Script (TensorFlow) amazon-sagemaker-workshop/TensorFlow_Distributed_MNIST.ipynb Deep Learning Framework: TensorFlow Neural Network Model: Bring your own network Dataset: MNIST Training Mode: Distributed Training with 2 or more nodes model_fn MNIST Training Data mnist.py Defines network architecture Main script used for training and hosting • model_fn : a function that contains all the logic to support training, evaluation, and prediction • train_input_fn : preprocess and load training data • eval_input_fn : preprocess and load evaluation data • serving_input_fn : defines the features to be passed to the model during prediction
  • 45. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Bring Your Own Script (TensorFlow) amazon-sagemaker-examples/sagemaker-python-sdk/tensorflow_resnet_cifar10_with_tensorboard Deep Learning Framework: TensorFlow Neural Network Model: Bring your own model Dataset: MNIST Training Mode: Distributed Training with 2 or more nodes MNIST Training Data Training at SageMaker Model Artifact Deployment at SageMaker Amazon SageMaker Mnist.py Amazon ECR
  • 46. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved.
  • 47. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. Bring Your Own Container (Any ML/DL Framework)
  • 48. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Bring Your Own Container (R) sample-notebooks/advanced_functionality/r_bring_your_own Any Machine Learning or Deep Learning framework Any Programming Language mars.R plumber.R Dockerfile Functions for training (fit) or serving the model Using plumber package to create a lightweight HTTP server for serving Defining a Docker container image for Training and Hosting
  • 49. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Bring Your Own Container (R) sample-notebooks/advanced_functionality/r_bring_your_own Use the same method for training and hosting model mars.R plumber.R Dockerfile Build&PushDocker Image Amazon ECR Training Data Training at SageMaker Model Artifact Deployment at SageMaker Amazon SageMaker
  • 50. © 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved. Hyper-Parameter Optimization (HPO)
  • 51. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Tuning Job Parameters { "Strategy": "Line"|"Random"|"Fill”|"Bayesian", "TuningJobObjective": { "Type": "Maximize"|"Minimize", "MetricName": "string" }, "ResourceLimits": { "MaxNumberOfTrainingJobs": integer, "MaxParallelTrainingJobs": integer, "MaxClockTimeInHours": integer }, "ParameterRanges": {...} }
  • 52. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Grid Search vs. Random Search Core illustration from Random Search for Hyper-Parameter Optimization by Bergstra and Bengio.
  • 53. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Random Search vs. Fill Search
  • 54. © 2018, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Bayesian Search