SlideShare a Scribd company logo
1 of 48
Download to read offline
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Industrialize Machine Learning
Using CI/CD Techniques
Felix Candelario
Principal Global Accounts
Solutions Architect
AWS
F S V 3 0 4 - i
Harsha Sharma
Senior Technical Account Manager
AWS
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Agenda
The need for governing ML workflows
A typical ML workflow
Solution
Demo
Sample audit
Remarks
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Highly regulated
workloads
Marketplace Lenders
• Operating exclusively online
• Niche product focus
• User of non-traditional data sources
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Strong controls
required
Control objectives
• Govern the process while
maintaining flexibility
• Same inputs should result in same
outputs
• Significantly improve auditability
Risks
• Potentially greater exposure to bias
• Reproducibility is hard
• Field in great state of flux
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
A typical ML workflow
On-Premise
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
A typical ML workflow
Client
On-Premise
Idea
Network
drive
Source code
v13
GPU Cluster
Source code
v13
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
A typical ML workflow
Client
On-Premise
Idea
Network
drive
GPU Cluster
Source code
v13
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
A typical ML workflow
Client
On-Premise
Idea
Network
drive
GPU Cluster
Source code
v1v1
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
A typical ML workflow
Client
On-Premise
Idea
Network
drive
GPU Cluster
Source code
v1
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
A typical ML workflow
Client
On-Premise
Idea
Network
drive
Model weightGPU Cluster Training job
Source code
v1
Source code
v2
Training job
v1.1
Training job
Error prone
• Manual orchestration
• Inputs aren’t versions
• Dataset
• Source code
• Outputs aren’t versions
• Weights
• Gathering evidence is manual
• Who made the change?
• When was the change made?
• What version of data was used?
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Solution
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Solution
Region
AWS Cloud
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Solution
Region
AWS Cloud
Source BucketBucket
Container
Training job
Input
Output
Dataset
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Solution
Region
AWS Cloud
BucketBucket
Container
Training job
Input
Output
Solution properties
• Flexible
• Collaborative
• Repeatable
• Auditable
• Secure
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
• Dockerfile
• manifest.json
• train
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
• Dockerfile
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
• Dockerfile
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
• Dockerfile
• manifest.json
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
• Dockerfile
• manifest.json
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source
Input
Output
Dataset
Contents of repo
• buildspec.yaml
• Dockerfile
• manifest.json
• train
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
Source BucketBucket
Input
Output
Dataset
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
BucketBucket
Container
Input
Output
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
BucketBucket
Container
Training job
Input
Output
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Demo
Region
AWS Cloud
BucketBucket
Container
Training job
Input
Output
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Sample audit
Region
AWS Cloud
Endpoints Inference
table
Apply for
loan
Trained model
Subsystem 1
Subsystem 2
Subsystem 3
Subsystem 4
Sample questions
• What version of the model produced
this inference?
• Who made the latest change to this
model?
• What version of training data was
exposed to this model?
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Sample audit
Region
AWS Cloud
Endpoints Inference
table
Apply for
loan
Trained model
Subsystem 1
Subsystem 2
Subsystem 3
Subsystem 4
Answers an API call away
sagemaker_client.describe_endpoint()
sagemaker_client.describe_endpoint_config
sagemaker_client.describe_model()
sagemaker_client.list_training_jobs()
codecommit_client.get_commit()
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Remarks
Region
AWS Cloud
BucketBucket
Container
Training job
Input
Output
Solution properties
• Flexible
• Collaborative
• Repeatable
• Auditable
• Secure
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Related breakouts
Thursday, Nov 29
Build, Train, and Deploy ML Models Quickly and Easily with Amazon SageMaker, ft. Intuit
4:00 PM – 5:00 PM | Venetian, Level 3, San Polo 3405
Thursday, Nov 29
Build, Train, and Deploy ML Models with Amazon SageMaker
12:15 PM – 2:30 PM | Bellagio, Level 1, Grand Ballroom 9
Thank you!
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Felix Candelario
fcandela@amazon.com
© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

More Related Content

What's hot

Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018
Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018
Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018Amazon Web Services
 
How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...
How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...
How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...Amazon 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
 
Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018
Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018
Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018Amazon Web Services
 
Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...
Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...
Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...Amazon Web Services
 
Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...
Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...
Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...Amazon Web Services
 
Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...
Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...
Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...Amazon Web Services
 
Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...
Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...
Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...Amazon Web Services
 
Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...
Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...
Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...Amazon Web Services
 
[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018
[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018
[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018Amazon Web Services
 
Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...
Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...
Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...Amazon Web Services
 
Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018
Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018
Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018Amazon Web Services
 
Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...
Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...
Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...Amazon Web Services
 
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
 
Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018
Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018
Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018Amazon Web Services
 
High Performance Computing on AWS: Driving Innovation without Infrastructure ...
High Performance Computing on AWS: Driving Innovation without Infrastructure ...High Performance Computing on AWS: Driving Innovation without Infrastructure ...
High Performance Computing on AWS: Driving Innovation without Infrastructure ...Amazon Web Services
 
Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...
Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...
Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...Amazon Web Services
 
What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018
What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018
What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018Amazon Web Services
 
How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018
How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018
How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018Amazon Web Services
 
Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018
Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018
Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018Amazon Web Services
 

What's hot (20)

Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018
Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018
Breaking Up the Monolith While Migrating to AWS (GPSTEC320) - AWS re:Invent 2018
 
How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...
How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...
How Verizon is Accelerating Cloud Adoption and Migration with the AWS Service...
 
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...
 
Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018
Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018
Foundations of AWS Global Cloud Infrastructure (ARC217) - AWS re:Invent 2018
 
Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...
Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...
Serverless Video Ingestion & Analytics with Amazon Kinesis Video Streams (ANT...
 
Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...
Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...
Automate & Audit Cloud Governance & Compliance in Your Landing Zone (ENT315-R...
 
Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...
Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...
Build an AWS Analytics Solution to Monitor the Video Streaming Experience (MA...
 
Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...
Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...
Configuration Management and Service Discovery with AWS Lambda (SRV338-R1) - ...
 
Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...
Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...
Bring the Power of AI to Your Amazon Connect Contact Center (BAP322-R1) - AWS...
 
[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018
[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018
[NEW LAUNCH!] Introduction to AWS Security Hub (SEC397) - AWS re:Invent 2018
 
Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...
Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...
Deploying Your ONNX Deep Learning with Apache MXNet Model Server (AIM413) - A...
 
Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018
Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018
Operationalizing Microsoft Workloads (WIN320) - AWS re:Invent 2018
 
Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...
Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...
Computing at the Edge with AWS Greengrass and Amazon FreeRTOS, ft. General El...
 
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...
 
Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018
Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018
Serverless:It All Started in Vegas (DVC306) - AWS re:Invent 2018
 
High Performance Computing on AWS: Driving Innovation without Infrastructure ...
High Performance Computing on AWS: Driving Innovation without Infrastructure ...High Performance Computing on AWS: Driving Innovation without Infrastructure ...
High Performance Computing on AWS: Driving Innovation without Infrastructure ...
 
Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...
Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...
Reserve Amazon EC2 On-Demand Capacity for Any Duration with On-Demand Capacit...
 
What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018
What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018
What's New with Amazon Redshift ft. McDonald's (ANT350-R1) - AWS re:Invent 2018
 
How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018
How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018
How Cox Automotive Runs GitHub Enterprise on AWS (ENT356-S) - AWS re:Invent 2018
 
Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018
Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018
Improve Accessibility Using Machine Learning (AIM332) - AWS re:Invent 2018
 

Similar to Industrialize ML with CI/CD

Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...
Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...
Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...Amazon Web Services
 
Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...
Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...
Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...Amazon Web Services
 
AWSome Day Online Conference 2018 Module 1.pdf
AWSome Day Online Conference 2018 Module 1.pdfAWSome Day Online Conference 2018 Module 1.pdf
AWSome Day Online Conference 2018 Module 1.pdfAmazon Web Services
 
PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...
PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...
PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...Amazon Web Services
 
Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...
Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...
Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...Amazon Web Services
 
Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...
Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...
Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...Amazon Web Services
 
CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018
CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018
CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018Bhuvaneswari Subramani
 
Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...
Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...
Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...Amazon Web Services
 
Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...
Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...
Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...Amazon Web Services
 
CI/CD pipelines on AWS - Builders Day Israel
CI/CD pipelines on AWS - Builders Day IsraelCI/CD pipelines on AWS - Builders Day Israel
CI/CD pipelines on AWS - Builders Day IsraelAmazon Web Services
 
Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...
Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...
Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...Amazon Web Services
 
[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...
[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...
[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...Amazon Web Services
 
Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...
Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...
Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...Amazon Web Services
 
CI/CD for AWS Lambda Projects - IsraelCloud Meetup
CI/CD for AWS Lambda Projects - IsraelCloud MeetupCI/CD for AWS Lambda Projects - IsraelCloud Meetup
CI/CD for AWS Lambda Projects - IsraelCloud MeetupBoaz Ziniman
 
Hybrid Cloud Customer Use Cases on AWS
Hybrid Cloud Customer Use Cases on AWSHybrid Cloud Customer Use Cases on AWS
Hybrid Cloud Customer Use Cases on AWSTom Laszewski
 
Ci/CD for AWS Lambda Projects - JLM CTO Club
Ci/CD for AWS Lambda Projects - JLM CTO ClubCi/CD for AWS Lambda Projects - JLM CTO Club
Ci/CD for AWS Lambda Projects - JLM CTO ClubBoaz Ziniman
 
Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...
Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...
Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...Amazon Web Services
 
Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...
Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...
Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...Amazon Web Services
 
[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...
[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...
[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...Amazon Web Services
 
Build CICD Pipeline for Container Presentation Slides
Build CICD Pipeline for Container Presentation SlidesBuild CICD Pipeline for Container Presentation Slides
Build CICD Pipeline for Container Presentation SlidesAmazon Web Services
 

Similar to Industrialize ML with CI/CD (20)

Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...
Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...
Set Up a CI/CD Pipeline for Deploying Containers Using the AWS Developer Tool...
 
Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...
Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...
Remove Undifferentiated Heavy Lifting from CI/CD Toolsets with Corteva Agrisc...
 
AWSome Day Online Conference 2018 Module 1.pdf
AWSome Day Online Conference 2018 Module 1.pdfAWSome Day Online Conference 2018 Module 1.pdf
AWSome Day Online Conference 2018 Module 1.pdf
 
PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...
PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...
PaaS – From Code to Running Application using AWS Elastic Beanstalk (DEV323) ...
 
Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...
Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...
Scaling and Automating DevOps with CloudBees and Spot Instances (GPSTEC310) -...
 
Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...
Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...
Build Modern Applications that Align with Twelve-Factor Methods (API303) - AW...
 
CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018
CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018
CI CD using AWS Developer Tools @ AWS Community Day Bengaluru 2018
 
Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...
Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...
Making Hybrid Work for You: Getting into the Cloud Fast (GPSTEC308) - AWS re:...
 
Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...
Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...
Enabling Your Organization’s Amazon Redshift Adoption – Going from Zero to He...
 
CI/CD pipelines on AWS - Builders Day Israel
CI/CD pipelines on AWS - Builders Day IsraelCI/CD pipelines on AWS - Builders Day Israel
CI/CD pipelines on AWS - Builders Day Israel
 
Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...
Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...
Accelerate Digital Transformation for Telecom Operators with Cloud-Native Amd...
 
[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...
[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...
[REPEAT 1] Safeguard the Integrity of Your Code for Fast and Secure Deploymen...
 
Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...
Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...
Safeguard the Integrity of Your Code for Fast and Secure Deployments (DEV349-...
 
CI/CD for AWS Lambda Projects - IsraelCloud Meetup
CI/CD for AWS Lambda Projects - IsraelCloud MeetupCI/CD for AWS Lambda Projects - IsraelCloud Meetup
CI/CD for AWS Lambda Projects - IsraelCloud Meetup
 
Hybrid Cloud Customer Use Cases on AWS
Hybrid Cloud Customer Use Cases on AWSHybrid Cloud Customer Use Cases on AWS
Hybrid Cloud Customer Use Cases on AWS
 
Ci/CD for AWS Lambda Projects - JLM CTO Club
Ci/CD for AWS Lambda Projects - JLM CTO ClubCi/CD for AWS Lambda Projects - JLM CTO Club
Ci/CD for AWS Lambda Projects - JLM CTO Club
 
Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...
Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...
Ensuring Your Windows Server Workloads Are Well-Architected - AWS Online Tech...
 
Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...
Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...
Control for Your Cloud Environment Using AWS Management Tools (ENT226-R1) - A...
 
[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...
[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...
[NEW LAUNCH!] Advancing Software Procurement in a Containerized World with th...
 
Build CICD Pipeline for Container Presentation Slides
Build CICD Pipeline for Container Presentation SlidesBuild CICD Pipeline for Container Presentation Slides
Build CICD Pipeline for Container Presentation Slides
 

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
 

Industrialize ML with CI/CD

  • 1.
  • 2. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Industrialize Machine Learning Using CI/CD Techniques Felix Candelario Principal Global Accounts Solutions Architect AWS F S V 3 0 4 - i Harsha Sharma Senior Technical Account Manager AWS
  • 3. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Agenda The need for governing ML workflows A typical ML workflow Solution Demo Sample audit Remarks
  • 4. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 5. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Highly regulated workloads Marketplace Lenders • Operating exclusively online • Niche product focus • User of non-traditional data sources
  • 6. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Strong controls required Control objectives • Govern the process while maintaining flexibility • Same inputs should result in same outputs • Significantly improve auditability Risks • Potentially greater exposure to bias • Reproducibility is hard • Field in great state of flux
  • 7. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 8. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. A typical ML workflow On-Premise
  • 9. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. A typical ML workflow Client On-Premise Idea Network drive Source code v13 GPU Cluster Source code v13
  • 10. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. A typical ML workflow Client On-Premise Idea Network drive GPU Cluster Source code v13
  • 11. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. A typical ML workflow Client On-Premise Idea Network drive GPU Cluster Source code v1v1
  • 12. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. A typical ML workflow Client On-Premise Idea Network drive GPU Cluster Source code v1
  • 13. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. A typical ML workflow Client On-Premise Idea Network drive Model weightGPU Cluster Training job Source code v1 Source code v2 Training job v1.1 Training job Error prone • Manual orchestration • Inputs aren’t versions • Dataset • Source code • Outputs aren’t versions • Weights • Gathering evidence is manual • Who made the change? • When was the change made? • What version of data was used?
  • 14. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 15. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Solution
  • 16. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Solution Region AWS Cloud
  • 17. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Solution Region AWS Cloud Source BucketBucket Container Training job Input Output Dataset
  • 18. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Solution Region AWS Cloud BucketBucket Container Training job Input Output Solution properties • Flexible • Collaborative • Repeatable • Auditable • Secure
  • 19. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 20. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset
  • 21.
  • 22.
  • 23. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml • Dockerfile • manifest.json • train
  • 24. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml
  • 25. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml
  • 26. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml • Dockerfile
  • 27. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml • Dockerfile
  • 28. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml • Dockerfile • manifest.json
  • 29. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml • Dockerfile • manifest.json
  • 30. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source Input Output Dataset Contents of repo • buildspec.yaml • Dockerfile • manifest.json • train
  • 31. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud Source BucketBucket Input Output Dataset
  • 32.
  • 33. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud BucketBucket Container Input Output
  • 34.
  • 35. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud BucketBucket Container Training job Input Output
  • 36.
  • 37.
  • 38.
  • 39. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Demo Region AWS Cloud BucketBucket Container Training job Input Output
  • 40. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 41. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Sample audit Region AWS Cloud Endpoints Inference table Apply for loan Trained model Subsystem 1 Subsystem 2 Subsystem 3 Subsystem 4 Sample questions • What version of the model produced this inference? • Who made the latest change to this model? • What version of training data was exposed to this model?
  • 42.
  • 43. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Sample audit Region AWS Cloud Endpoints Inference table Apply for loan Trained model Subsystem 1 Subsystem 2 Subsystem 3 Subsystem 4 Answers an API call away sagemaker_client.describe_endpoint() sagemaker_client.describe_endpoint_config sagemaker_client.describe_model() sagemaker_client.list_training_jobs() codecommit_client.get_commit()
  • 44. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.
  • 45. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Remarks Region AWS Cloud BucketBucket Container Training job Input Output Solution properties • Flexible • Collaborative • Repeatable • Auditable • Secure
  • 46. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Related breakouts Thursday, Nov 29 Build, Train, and Deploy ML Models Quickly and Easily with Amazon SageMaker, ft. Intuit 4:00 PM – 5:00 PM | Venetian, Level 3, San Polo 3405 Thursday, Nov 29 Build, Train, and Deploy ML Models with Amazon SageMaker 12:15 PM – 2:30 PM | Bellagio, Level 1, Grand Ballroom 9
  • 47. Thank you! © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Felix Candelario fcandela@amazon.com
  • 48. © 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.