Weitere ähnliche Inhalte Ähnlich wie Build, train and deploy machine learning models at scale using AWS (20) Mehr von Amazon Web Services (20) Build, train and deploy machine learning models at scale using AWS1. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Build, train and deploy Machine Learning
models on Amazon SageMaker
Julien Simon
Global Evangelist, AI & Machine Learning
@julsimon
Nils Mohr
Flight Data Programmer Analyst
British Airways
2. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Amazon SageMaker
1
2
3
3. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Build, train and deploy models using SageMaker
Business
Problem
ML problem framing Data collection
Data integration
Data preparation and
cleaning
Data visualization
and analysis
Feature engineering
Model training and
parameter tuning
Model evaluation
Monitoring and
debugging
Model deployment
Predictions
Are business
goals
met?
YESNO
Dataaugmentation
Feature
augmentation
Re-training
Neo
Elastic inference
Ground Truth
4. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Aircraft Information Intelligence
British Airways Predictive Maintenance Platform
Nils Mohr
Flight Data Programmer Analyst
British Airways
S U M M I T
5. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
British Airways
• 290+ aircraft
• 46 million+ passengers per year
• 200+ destinations in 75+ countries
S U M M I T
6. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
7. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
What is flight data?
8. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Timeseries data
9. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Timeseries data
10. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Report data
11. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Our dataset
12. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Reasons for the development
13. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
14. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
High level architecture
British Airways
Storage in
Amazon S3
AWS Cloud
On prem software/tools
Transfer into AWS
15. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Our AWS Architecture
AWS Cloud
Amazon VPC
Storage in
Amazon S3
Data optimization
and cleaning on
Amazon Fargate
Config and Metadata store
Amazon
SQS Metadata
generation on
Amazon Fargate
Amazon
SQS
Problem evaluation
on AWS Lambda
Amazon
SQS
16. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Traditional Event detection
AWS Cloud
Amazon VPC
Amazon SQS
Config and Metadata store
Lambda function
checks config
and sends SQS messages
Lambda function
runs custom/bespoke script
and creates an alert if necessary
Amazon SQS
17. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Event detection with Amazon SageMaker
AWS Cloud
Amazon VPC
Amazon SQS
Config and Metadata store
Lambda function
checks config
and creates endpoints
Amazon SageMaker models
trained on specific features
Amazon SageMaker
endpoint
Lambda function
creates dataset for inference
Amazon SQS
Amazon SQS
Lambda function evaluates
data
and create alerts
if needed
18. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Lessons learned
Understand the problem
Ensure data quality
Ask questions
Apply the easiest solution
19. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
20. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Where do we go from here?
Aircraft
Engineers
Create a training job in
Amazon SageMaker
Resulting model
21. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Where do we go from here?
Aircraft
Engineers
22. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
23. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Model options
Training code
Factorization Machines
Linear Learner
Principal Component Analysis
K-Means Clustering
XGBoost
And more
Built-in Algorithms (17)
No ML coding required
No infrastructure work required
Distributed training
Pipe mode
Bring Your Own Container
Full control, run anything!
R, C++, etc.
No infrastructure work required
Built-in Frameworks
Bring your own code: script mode
Open source containers
No infrastructure work required
Distributed training
Pipe mode
24. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Built-in Deep Learning frameworks: just add your code
• Built-in containers for training and prediction.
• Code available on Github, e.g. https://github.com/aws/sagemaker-tensorflow-containers
• Build them, run them on your own machine, customize them, etc.
• Script mode: use the same code as on your laptop
No infrastructure work required: simply define instance type and instance count
Distributed training out of the box: zero setup
Pipe mode: stream infinitely large datasets directly from Amazon S3
25. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
AWS: The platform of choice to run TensorFlow
85% of all
TensorFlow
workloads in the
cloud runs on AWS
26. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Training ResNet-50 with the
ImageNet dataset using our
optimized build of Tensorflow 1.11 on
a c5.18xlarge instance type is 11x
faster than training on the stock
binaries.
Optimizing Tensorflow for Amazon EC2 instances
C5 instances (Intel Skylake)
65%
90%
P3 instances (NVIDIA V100)
27. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Demo: Keras+Tensorflow
Script mode
Automatic model tuning
Elastic inference
https://gitlab.com/juliensimon/dlnotebooks/tree/master/keras/04-fashion-
mnist-sagemaker-advanced
28. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
ApacheMXNet:DeepLearningforenterprisedevelopers
• Gluon CV Gluon NLP
ONNX
2x faster
Java Scala
29. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
Demo: Gluon CV
State of the art models in just a few lines of code
https://gluon-cv.mxnet.io/
30. © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.S U M M I T
© 2018, Amazon Web Services, Inc. or Its Affiliates. All rights reserved.
Getting started
http://aws.amazon.com/free
https://aws.amazon.com/sagemaker
https://github.com/aws/sagemaker-python-sdk
https://github.com/awslabs/amazon-sagemaker-examples
https://medium.com/@julsimon
https://gitlab.com/juliensimon/dlnotebooks
31. Thank you!
S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Julien Simon
Global Evangelist, AI and Machine Learning
@julsimon
32. S U M M I T © 2019, Amazon Web Services, Inc. or its affiliates. All rights reserved.