OML Usage Highlight: ML on SailGP data: Predicting the best sailing direction

Slide 1

Slide 1 text

OML usage highlight: ML on SailGP data: Predicting the best sailing direction OML Office Hours Jeroen Kloosterman Product Strategy Director SMB EMEA, Analytics and AI Supported by Marcos Arancibia, Mark Hornick and Sherry LaMonica Product Management, Oracle Machine Learning Move the Algorithms; Not the Data! Copyright © 2021, Oracle and/or its affiliates. This Session will be Recorded

Slide 2

Slide 2 text

• Now Available DB21c: • OML4Py on-premise launch with DB 21c • New algorithms with OML4SQL: XGBoost and MSET • ML on SailGP data: Predicting the best sailing direction • Q&A Topics for today Copyright © 2021, Oracle and/or its affiliates 2

Slide 3

Slide 3 text

[email protected] Product Strategy Director SMB EMEA Analytics & Machine Learning Oracle Machine Learning AskTOM Office Hours ML on SailGP data: Predicting the best sailing direction

Slide 4

Slide 4 text

Sail GP – Sailing athletes + Data athletes

Slide 5

Slide 5 text

What is SailGP? What is Sail GP? • World’s fastest sailing yachts • Iconic venues around the world • 8 Teams, identical boats, all teams see all data • 3K+ data points from 8 yachts • Backbone is Oracle Cloud

Slide 6

Slide 6 text

What is SailGP? Products • Autonomous Datawarehouse • Oracle Analytics Cloud • Oracle Streaming Analytics

Slide 7

Slide 7 text

Analysis of Start of Race Analysis of Maneuvers Predict the best sailing direction What’s in the workshop?

Slide 8

Slide 8 text

Slide 9

Slide 9 text

Slide 10

Slide 10 text

Wind speed Wind angle 1 Boat speed 1 Wind angle 2 Boat speed 2

Slide 11

Slide 11 text

Slide 12

Slide 12 text

Copyright © 2021, Oracle and/or its affiliates. All rights reserved. Wind Angle Wind Speed Resulting Boat Speed 95.6 15.9 45.7 58.2 24.3 42.4 88.5 26.7 50.5 67.3 2.5 3.8 … … … Historic measurements Model Training algorithm 1. Train 2. Predict Prediction Wind Angle Wind Speed 42.1 22.7 Resulting Boat Speed 30.5

Slide 13

Slide 13 text

Products Autonomous Data Warehouse Oracle Machine Learning Oracle Analytics Cloud Data Team Analysts Umpires Audience Oracle Cloud Infrastructure (OCI)

Slide 14

Slide 14 text

Demonstration 1. Explore data in OAC 2. Build model in OML (AutoML) 3. Predict from OAC -> OML 4. Visualize results in OAC

Slide 15

Slide 15 text

Slide 16

Slide 16 text