Ready to use deep learning models

Slide 1

Slide 1 text

R E A D Y T O U S E D E E P L E A R N I N G M O D E L S —  Gabriela de Queiroz  Sr. Engineering & Data Science Manager, IBM Founder, R-Ladies Founder, AI Inclusive @gdequeiroz | www.k-roz.com Slides: http://bit.ly/max-gdgdevfest19

Slide 2

Slide 2 text

Center for Open Source Data and AI Technologies (CODAIT) - CODAIT aims to make AI solutions easier to create, deploy, and manage in the enterprise - Relaunch of the Spark Technology Center (STC) to reflect expanded mission Watson West Building 505 Howard St. San Francisco, California CODAIT codait.org @gdequeiroz | www.k-roz.com | [email protected]

Slide 3

Slide 3 text

30+ open source developers! • TensorFlow • PyTorch • Keras • Apache Arrow • ONNX • Jupyter • Apache Spark • Kubeflow • And more

Slide 4

Slide 4 text

Machine Learning Team (10 open source developers) 1) TensorFlow 2) PyTorch 3) Keras 4) Apache Arrow

Slide 5

Slide 5 text

Have you ever wanted to classify images, identify objects or generate captions of images? @gdequeiroz | www.k-roz.com | [email protected] Slides: http://bit.ly/max-gdgdevfest19

Slide 6

Slide 6 text

With the Model Asset eXchange, you can!

Slide 7

Slide 7 text

Applying Deep Learning: Perception Data ??? Train model ??? $$$ Get model ??? Deploy model ??? $$$ Training – Data Scientist Consumption – App Developer, Domain Expert Slides: http://bit.ly/max-gdgdevfest19

Slide 8

Slide 8 text

Applying Deep Learning: Reality Find model Get code Test, verify, fix Train model Deploy model Use model Discovery Execution Consumption 1 2 3 4A 4B 5 @gdequeiroz | www.k-roz.com | [email protected]

Slide 9

Slide 9 text

Step 1: Find a model … that does what you need … that is free to use … that is performant enough

Slide 10

Slide 10 text

Step 2: Get the code Is there a good implementation available? … that does what you need … that is free to use … that is performant enough TensorFlow code to build ResNet50 neural network graph

Slide 11

Slide 11 text

Or … Step 2: Get the pre-trained weights Is there a good pre-trained model available? … that does what you need … that is free to use … that is performant enough

Slide 12

Slide 12 text

Step 3: Verify the model you found Check … … that does what you need … that is free to use (license) … that is performant enough (computation & accuracy)

Slide 13

Slide 13 text

Step 4 (a): Train the model

Slide 14

Slide 14 text

Step 4 (a): Train the model

Slide 15

Slide 15 text

Step 4 (b): Figure out how to deploy your model … adjust inference code (or write from scratch) … package inference code and model code, and pre-trained weights together … deploy your package

Slide 16

Slide 16 text

Step 5: Consume your model … plug in into your application

Slide 17

Slide 17 text

Step 6: Proﬁt … hopefully

Slide 18

Slide 18 text

Applying Deep Learning: Reality Find model Get code Test, verify, fix Train model Deploy model Use model Discovery Execution Consumption 1 2 3

Slide 19

Slide 19 text

Model Asset eXchange • One-stop place for developers/data scientists to ﬁnd and use free and open source deep learning models ibm.biz/model-exchange @gdequeiroz | www.k-roz.com | [email protected]

Slide 20

Slide 20 text

Model Asset eXchange (MAX) • Wide variety of domains (text, audio, image, etc) • Multiple deep learning frameworks • Vetted and tested code and IP • Trainable and Deployable versions • Build and deploy a model web service in seconds ibm.biz/model-exchange

Slide 21

Slide 21 text

What do I need to get started?

Slide 22

Slide 22 text

No content

Slide 23

Slide 23 text

Find* a state-of-art open source deep learning model specific to domain Validate license terms Perform model health check & code clean up Wrap models in MAX framework and provide REST API Publish the deployable model as Docker images on Docker Hub Use the MAX training framework to create an image for custom model training Review and Continuous Integration * or build from scratch Model Asset eXchange (MAX) ibm.biz/model-exchange BEHIND THE SCENES

Slide 24

Slide 24 text

Ways of accessing the model

Slide 25

Slide 25 text

OBJECT DETECTOR Localize and identify multiple objects in a single image @gdequeiroz | www.k-roz.com | [email protected]

Slide 26

Slide 26 text

No content

Slide 27

Slide 27 text

No content

Slide 28

Slide 28 text

1) REST API MODEL REST API flask Request information from model Send input to model GET POST

Slide 29

Slide 29 text

1.1) Via Swagger

Slide 30

Slide 30 text

1.2) Via Python You can try! http://ibm.biz/max-notebook

Slide 31

Slide 31 text

1.2) Via R

Slide 32

Slide 32 text

2) WEB APP • User uses Web UI to send an image to Model API • Model API returns object data and Web UI displays detected objects •

Slide 33

Slide 33 text

2) WEB APP bit.ly/object-detector-app

Slide 34

Slide 34 text

3) Node-RED flow

Slide 35

Slide 35 text

4) CodePen

Slide 36

Slide 36 text

What if I want to train a model?

Slide 37

Slide 37 text

No content

Slide 38

Slide 38 text

No content

Slide 39

Slide 39 text

And there is more!

Slide 40

Slide 40 text

MAX-Framework MAX-Skeleton •A pip installable python library •Wrapper around flask •Abstracts out all basic functionality of the MAX model into MAXModelWrapper and MAXApi abstract classes github.com/IBM/MAX-Framework •Template to create a deployable MAX model •Contains all the code scaffolding and imports MAX Framework github.com/IBM/MAX-Skeleton

Slide 41

Slide 41 text

All that is available for YOU for FREE @gdequeiroz | www.k-roz.com | [email protected]

Slide 42

Slide 42 text

How do I get started? @gdequeiroz | www.k-roz.com | [email protected]

Slide 43

Slide 43 text

@gdequeiroz | www.k-roz.com | [email protected] bit.ly/max-tutorial

Slide 44

Slide 44 text

@gdequeiroz | www.k-roz.com | [email protected] Code Patterns How to easily consume MAX models bit.ly/max-code-patterns

Slide 45

Slide 45 text

@gdequeiroz | www.k-roz.com | [email protected]

Slide 46

Slide 46 text

Data Asset eXchange (DAX) • Curated free and open datasets under open data licenses • Standardized dataset formats and metadata • Ready for use in enterprise AI applications • Complement to the Model Asset eXchange (MAX) • ibm.biz/data-asset-exchange

Slide 47

Slide 47 text

Data Asset eXchange (DAX)

Slide 48

Slide 48 text

Data Asset eXchange (DAX)

Slide 49

Slide 49 text

Thank you! K- ROZ .COM @GDEQUEIROZ GABRI EL A DE QU EI ROZ ai-inclusive.org Our mission is to increase the representation and participation of gender minority groups in AI. [email protected] ibm.biz/model-exchange Slides: http://bit.ly/max-gdgdevfest19