Inside TensorFlow- Building ML Infra

Inside TensorFlow- Building ML Infra

TensorFlow via YouTube Direct link

Intro

1 of 26

1 of 26

Intro

Class Central Classrooms beta

YouTube playlists curated by Class Central.

Classroom Contents

Inside TensorFlow- Building ML Infra

Automatically move to the next video in the Classroom when playback concludes

  1. 1 Intro
  2. 2 Big data and ML infra are similar
  3. 3 Speaker background
  4. 4 Why invest in ML infra?
  5. 5 Case study: Building a new TF runtime
  6. 6 ML program as a computational graph
  7. 7 An example ML program
  8. 8 Lifetime of an ML program
  9. 9 Vectorized normalization
  10. 10 A slight digression on Eager execution
  11. 11 ML infra and SQL query processing
  12. 12 (Random) scan-based access patterns
  13. 13 Beyond pure dataflow
  14. 14 ML and DB terminology mapping
  15. 15 Recall graph processing workflow
  16. 16 Expressing input pipelines
  17. 17 Decoupled API and execution
  18. 18 Challenge: Randomized transformations
  19. 19 Graph rewrites
  20. 20 Cost model and data stats
  21. 21 Constraint propagation
  22. 22 Storage/access optimizations
  23. 23 Push vs pull based execution
  24. 24 Distributed and parallel execution
  25. 25 ML infra is like data infra, with new twists
  26. 26 Let's collaborate

Never Stop Learning.

Get personalized course recommendations, track subjects and courses with reminders, and more.

Someone learning on their laptop while sitting on the floor.