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 for Vehicle Price Prediction - Delivered Continuously at AutoScout24

332 views

Published on

https://predictiveanalyticsworld.de/de/berlin2017/programm/#session37621

AutoScout24 is the largest online car marketplace Europe-wide. With more than 2.4 million listings across Europe, AutoScout24 has access to large amounts of data about historic and current market prices and wants to use this data to empower its users to make informed decisions about selling and buying cars. We created a live price estimation service for used vehicles based on a Random Forest prediction model that is continuously delivered to the end user. Learn how automated verification using live test data sets in our delivery pipeline allows us to release model improvements with confidence at any time.

Published in: Internet
  • Be the first to comment

Predictive Analytics for Vehicle Price Prediction - Delivered Continuously at AutoScout24

  1. 1. Predictive Analytics World Berlin | 14.11.2017 | Christian Deger | @cdeger Predictive Analytics for Vehicle Price Prediction Delivered Continuously at AutoScout24
  2. 2. Christian Deger Chief Architect christian@deger.eu @cdeger
  3. 3. 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
  4. 4. The task: A consumer-facing data product
  5. 5. The task: A consumer-facing data product
  6. 6. The task: A consumer-facing data product
  7. 7. The prediction model Car listings of last two years
  8. 8. The prediction model Car listings of last two years
  9. 9. The prediction model: Random forest Volkswagen GolfCar listings of last two years
  10. 10. How to turn an R-based prediction model into a high-performance web application? ?
  11. 11. How to turn an R-based prediction model into a high-performance web application?
  12. 12. How to turn an R-based prediction model into a high-performance web application?
  13. 13. How to turn an R-based prediction model into a high-performance web application?  Continuous Delivery!
  14. 14. Application code in one repository per service. Typical delivery pipeline
  15. 15. Application code in one repository per service. CI Deployment package as artifact. Typical delivery pipeline
  16. 16. Application code in one repository per service. CI Deployment package as artifact. CD Deliver package to servers Typical delivery pipeline
  17. 17. Continuous delivery pipelines Prediction Model Pipeline
  18. 18. Continuous delivery pipelines Prediction Model Pipeline Web Application Pipeline
  19. 19. The price for CD: Extensive model validation
  20. 20. The price for CD: Extensive model validation
  21. 21. Lessons learned Form a cross-functional team of data scientists & software engineers! Software engineers … learn how data scientists work … and understand the quirks of a prediction model Data Scientist … learn about unit testing, stable interfaces, git, etc. ... get quick feedback about the impact of their work  Model and product iterations become much faster!
  22. 22. Lessons learned Generating gigabytes of Java code is a challenge for the JVM Use the G1 garbage collector Turn off Tiered Compilation  Do extensive warm-ups
  23. 23. Lessons learned – Warm up
  24. 24. Lessons learned The approach of applying Continuous Delivery to Data Science is useful independently of the tech  Successfully applied similarly to a Python- and Spark-based project  Even more useful when quick model evolution is required because of rapidly changing inputs (e.g. user interaction)
  25. 25. Conclusions  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.
  26. 26. Conclusions: Price evaluation everywhere
  27. 27. Conclusions: Price evaluation everywhere
  28. 28. Conclusions: Price evaluation everywhere
  29. 29. Conclusions: Price evaluation everywhere

×