Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.
Predictive Analytics World London| Oct 12, 2016 | Arif Wider & Christian Deger
Predictive Analytics for Vehicle Price Pred...
Dr. Arif Wider
Senior Consultant/Developer
awider@thoughtworks.com
@arifwider
Christian Deger
Chief Architect
cdeger@autoscout24.com
@cdeger
PL
S
RUS
UA
RO
CZ
D
NL
B
F
A
HR
I
E
BG
TR
18countries
2.4m+cars & motos
10m+users per
month
The task: A consumer-facing data product
5Predictive Analytics World London 2016 Predictive Analytics Delivered Continuous...
The task: A consumer-facing data product
6Predictive Analytics World London 2016 Predictive Analytics Delivered Continuous...
The task: A consumer-facing data product
7Predictive Analytics World London 2016 Predictive Analytics Delivered Continuous...
The prediction model: Random forest
8
Volkswagen GolfCar listings of
last two years
Predictive Analytics World London 2016...
How to turn an R-based prediction model
into a high-performance web application?
9
?
Predictive Analytics World London 201...
Traditional approach: Manually translate
model to an efficient implementation
10Predictive Analytics World London 2016 Pre...
Traditional approach: Manually translate
model to an efficient implementation again
11Predictive Analytics World London 20...
Our approach: Automatically generate
implementation & deliver continuously
12
 Continuous Delivery!
Predictive Analytics ...
Application code in
one repository per
service.
CI
Deployment package
as artifact.
CD
Deliver package to
servers
Predictiv...
Continuous delivery pipelines
14
Prediction Model Pipeline
Web Application Pipeline
Predictive Analytics World London 2016...
The price for CD: Extensive model validation
15
Lessons learned
16
 Form cross-functional teams of
data scientists & software engineers!
 Set up usage reporting early t...
Conclusions
17
 Continuous Delivery allows us to bring prediction
model changes live very quickly.
 Only extensive autom...
Thank you!
Questions?
– Arif Wider & Christian Deger –
Upcoming SlideShare
Loading in …5
×

Predictive Analytics for Vehicle Price Prediction - Delivered Continuously at AutoScout24

366 views

Published on

How we applied continuous delivery to data science to create a high-performance & quickly evolving data product. Presented at Predictive Analytics World Business London 2016 by Arif Wider (ThoughtWorks) and Christian Deger (AutoScout24).

Published in: Data & Analytics
  • Be the first to comment

  • Be the first to like this

Predictive Analytics for Vehicle Price Prediction - Delivered Continuously at AutoScout24

  1. 1. Predictive Analytics World London| Oct 12, 2016 | Arif Wider & Christian Deger Predictive Analytics for Vehicle Price Prediction Delivered Continuously at AutoScout24
  2. 2. Dr. Arif Wider Senior Consultant/Developer awider@thoughtworks.com @arifwider
  3. 3. Christian Deger Chief Architect cdeger@autoscout24.com @cdeger
  4. 4. PL S RUS UA RO CZ D NL B F A HR I E BG TR 18countries 2.4m+cars & motos 10m+users per month
  5. 5. The task: A consumer-facing data product 5Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  6. 6. The task: A consumer-facing data product 6Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  7. 7. The task: A consumer-facing data product 7Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  8. 8. The prediction model: Random forest 8 Volkswagen GolfCar listings of last two years Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  9. 9. How to turn an R-based prediction model into a high-performance web application? 9 ? Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  10. 10. Traditional approach: Manually translate model to an efficient implementation 10Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  11. 11. Traditional approach: Manually translate model to an efficient implementation again 11Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger :-/
  12. 12. Our approach: Automatically generate implementation & deliver continuously 12  Continuous Delivery! Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  13. 13. Application code in one repository per service. CI Deployment package as artifact. CD Deliver package to servers Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger Typical continuous delivery pipeline
  14. 14. Continuous delivery pipelines 14 Prediction Model Pipeline Web Application Pipeline Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  15. 15. The price for CD: Extensive model validation 15
  16. 16. Lessons learned 16  Form cross-functional teams of data scientists & software engineers!  Set up usage reporting early to improve your data product in a data-driven way.  Generating gigabytes of Java code is a challenge for the JVM  Use the G1 garbage collector  Do extensive warm-ups Turn off Tiered Compilation Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  17. 17. Conclusions 17  Continuous Delivery allows us to bring prediction model changes live very quickly.  Only extensive automated end-to-end tests provide confidence to deploy to production automatically.  Java code generation allows for very low response times and excellent scalability for high loads but requires plenty of memory. Predictive Analytics World London 2016 Predictive Analytics Delivered Continuously – A. Wider & C. Deger
  18. 18. Thank you! Questions? – Arif Wider & Christian Deger –

×