building and running your first machine learning...

© 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved.

Constantin Gonzalez

Principal Solutions Architect, Amazon Web Services

Building and running your first Machine

Learning application on Amazon

SageMaker


What you’ll get out of this session

• An overview of the Machine Learning (ML) process

• An overview of Amazon SageMaker

• Examples for:

• Using Jupyter Notebooks

• Feature extraction and data preparation in Python

• ML algorithms available in Amazon SageMaker

• Building, training and deploying ML models

• Hands-on

• Opportunities for Q&A


Overview


Long history of ML at Amazon

Personalized

recommendations

Inventing

entirely new

customer

experiences

Fulfillment

automation and

inventory

management

Drones Voice-driven

interactions


Machine Learning at AWS

Our mission:

Put machine learning in the hands of every developer and

data scientist


The Amazon Machine Learning stack

Frameworks & Infrastructure

Platform Services

Application Services


Bottom layer: frameworks & interfaces

NVIDIA

Tesla V100 GPUs

P3

AWS Deep Learning AMI

5,120 tensor cores

128 GB of memory

1 petaflop of compute

NVLink 2.0

~14X faster than P2



Frameworks & Interfaces

Platform Services


Caffe2 CNTKApache

MXNetPyTorch TensorFlow Torch Keras Gluon

AWS Deep Learning AMIs


Amazon SageMaker

End-to-End

Machine Learning

Platform

Zero setup Flexible Model

Training

Pay by the second

$

Build, train, and deploy machine learning models at scale




Platform Services


Caffe2 CNTKApache



Amazon SageMaker AWS DeepLens


Amazon ML application services

Vision Speech Language

Amazon Rekognition

Amazon Rekognition

Video

Amazon Polly

Amazon Transcribe

Amazon Lex

Amazon Translate

Amazon Comprehend




Platform Services


Caffe2 CNTKApache



Amazon SageMaker AWS DeepLens

Amazon

RekognitionAmazon Polly

Amazon

TranscribeAmazon Lex

Amazon

Translate

Amazon

Comprehend


Data Visualization &

Analysis

Business Problem –

ML problem framing Data Collection

Data Integration

Data Preparation &

Cleaning

Feature Engineering

Model Training &

Parameter Tuning

Model Evaluation

Are Business

Goals met?

Model Deployment

Monitoring &

Debugging

– Predictions

YesNo

Data

Au

gm

en

tati

on

Featu

re

Au

gm

en

tati

on

Re-training

The Machine Learning process


End-to-End

Machine Learning

Platform

Zero setup Flexible Model

Training

Pay by the second

$

Build, train, and deploy machine learning models at scale

Amazon SageMaker


Hands On!


Step 1: Create your notebook instance


Step 2: Load, visualize and prepare Data


Step 3: Train your model


Step 4: Inference


Next Steps

• Explore Amazon SageMaker built-in algorithms

• Build you ML endpoint into your web/mobile/IoT app

• Deploy your ML model to IoT devices using Amazon

Greengrass ML inference

• Collect new ground truth data from production

• Run A/B tests to find the best model

• Build automated data -> train -> deploy workflows

• Amazon CodePipeline, AWS Step Functions, etc.

• Use Amazon Mechanical Turk to help label your data

• Combine multiple ML models into an ML pipeline


Build something cool on Amazon SageMaker

• Getting started with Amazon SageMaker:

https://aws.amazon.com/sagemaker/

• Use the Amazon SageMaker SDK:

• For Python: https://github.com/aws/sagemaker-python-sdk

• For Spark: https://github.com/aws/sagemaker-spark

• SageMaker Examples: https://github.com/awslabs/amazon-sagemaker-

examples

• Let us know what you build!

https://github.com/aws/sagemaker-python-sdk


Please complete the session survey in

the summit mobile app.


Thank You!

building and running your first machine learning...

Documents