Tweet
Tweet

Slide 1

Slide 1 text

Hello World!

Slide 2

Slide 2 text

How’s the JOOOSHHH?

Slide 3

Slide 3 text

How’s the JOOOSHHH?

Slide 4

Slide 4 text

ML on GCP: Design-Develop-Deploy by @CharmiChokshi Machine Learning Engineer @Shipmnts.com

Slide 5

Slide 5 text

90% have failed to solve this...

Slide 6

Slide 6 text

99.9% have failed to solve this...

Slide 7

Slide 7 text

What we just did?

Slide 8

Slide 8 text

What we just did? We solved logical and complex* math puzzles in a fraction of time

Slide 9

Slide 9 text

Think about, How we solved them?

Slide 10

Slide 10 text

Think about, How we solved them? We had inputs We also had outputs

Slide 11

Slide 11 text

Think about, How we solved them? We created Rules and Patterns in our brains!!

Slide 12

Slide 12 text

Can machines create rules on their own?

Slide 13

Slide 13 text

Can machines create rules on their own? No, if it’s Classical Programming...

Slide 14

Slide 14 text

Can machines create rules on their own? YES, if it’s Machine Learning!!!

Slide 15

Slide 15 text

Classical Programming vs Machine Learning Classical Programming Machine Learning Rules Rules Data Data Answers Answers

Slide 16

Slide 16 text

ML Use Cases

Slide 17

Slide 17 text

Why so little ML Apps out there? ● Building and Scaling ML infrastructure is hard ● Operating production ML system is time consuming and Expensive

Slide 18

Slide 18 text

What if, you can get ● Fully managed service ● Training using custom tensorflow graph for any ML use cases ● Training at scale to shorten Dev Cycle ● Automatically maximize predictive accuracy with HyperTune ● Integrated Datalab experience

Slide 19

Slide 19 text

What if, you can get ● Fully managed service ● Taring using custom tensorflow graph for any ML use cases ● Training at scale to shorten Dev Cycle ● Automatically maximize predictive accuracy with HyperTune ● Integrated Datalab experience At a single place!!!

Slide 20

Slide 20 text

What if, you can get ● Fully managed service ● Taring using custom tensorflow graph for any ML use cases ● Training at scale to shorten Dev Cycle ● Automatically maximize predictive accuracy with HyperTune ● Integrated Datalab experience At a single place!!!

Slide 21

Slide 21 text

ML on Cloud Platform

Slide 22

Slide 22 text

ML on Cloud Platform

Slide 23

Slide 23 text

ML on Cloud Platform

Slide 24

Slide 24 text

ML on Cloud Platform

Slide 25

Slide 25 text

No content

Slide 26

Slide 26 text

ML on Cloud Platform DIY

Slide 27

Slide 27 text

Machine Learning as an API Access pre-trained models with a single REST API request

Slide 28

Slide 28 text

Available Pre-trained APIs

Slide 29

Slide 29 text

Available Pre-trained APIs

Slide 30

Slide 30 text

Available Pre-trained APIs

Slide 31

Slide 31 text

No content

Slide 32

Slide 32 text

Let’s identify series characters!

Slide 33

Slide 33 text

Let’s try the Vision API!

Slide 34

Slide 34 text

What if you want to train these APIs on custom data?

Slide 35

Slide 35 text

No content

Slide 36

Slide 36 text

Cloud AutoML Train custom models without writing a single line of code

Slide 37

Slide 37 text

The Aha! moment

Slide 38

Slide 38 text

How it works?

Slide 39

Slide 39 text

जीसीपी ही तसरा बाप हे, अहम् ब्रह्माि म!

Slide 40

Slide 40 text

No content

Slide 41

Slide 41 text

What if you want to train/deploy/scale your Keras based ML model on GCP? Custom task having custom model and data

Slide 42

Slide 42 text

What if your model has multiple inputs text and image? Can we get predictions for such cases using AutoML?

Slide 43

Slide 43 text

Cloud ML Engine Build, Train, and Serve custom models with your own data

Slide 44

Slide 44 text

Ways to create custom models on GCP

Slide 45

Slide 45 text

Some of the GCP offerings

Slide 46

Slide 46 text

Cloud ML Engine: IaaS

Slide 47

Slide 47 text

Where ML Engine fits in ML workflow

Slide 48

Slide 48 text

Components of AI Platform ● Training service ● Prediction service ● Notebooks ● Data labeling service (beta) ○ Submit a request to label your video, image, or text data with some instructions ● Deep learning VM image ● Tools to interact with AI Platform ○ Google Cloud Console ■ Stackdriver Logging ■ Stackdriver Monitoring ○ REST API

Slide 49

Slide 49 text

Design-Develop-Deploy

Slide 50

Slide 50 text

3 things from this session...

Slide 51

Slide 51 text

ML APIs

Slide 52

Slide 52 text

AutoML

Slide 53

Slide 53 text

Cloud ML Engine

Slide 54

Slide 54 text

Learn more...

Slide 55

Slide 55 text

Questions? Comments? Suggestions? Machine Learning on GCP: Design-Develop-Deploy by @CharmiChokshi

Slide 56

Slide 56 text

Thank you! Happy Learning :) Machine Learning on GCP: Design-Develop-Deploy by @CharmiChokshi