Completed
Reducing memory
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
TensorFlow Model Optimization - Quantization and Pruning
Automatically move to the next video in the Classroom when playback concludes
- 1 Introduction
- 2 Why is this important
- 3 Benefits of optimization
- 4 Resource constrained environment
- 5 Application constrained environment
- 6 Machine learning opportunities
- 7 Machine learning efficiency
- 8 Matrix multiply
- 9 Goals for optimization
- 10 Reducing precision
- 11 Reducing memory
- 12 Reducing bandwidth pressure
- 13 Reduce precision
- 14 Linear mapping
- 15 The problem
- 16 The implications
- 17 Quantization is complicated
- 18 Its hard to interpret
- 19 The model is not enough
- 20 Quantization types
- 21 Quantization benefits
- 22 Quantization tools
- 23 Posttraining
- 24 TensorFlow flowlight converter
- 25 Quantisation types
- 26 Highbury quantization
- 27 Accuracy
- 28 Interior Quantization
- 29 Results
- 30 Quantization training
- 31 Quantization model
- 32 Hybrid quantization
- 33 Integer quantization
- 34 Training scrape
- 35 Summary
- 36 Neural connection pruning
- 37 Stencil pruning
- 38 Tensor pruning
- 39 TensorFlow pruning API
- 40 Pruning schedule
- 41 Benefits of pruning
- 42 Roadmap
- 43 Better target hardware
- 44 Feedback
- 45 Tools
- 46 Questions
- 47 Training with integer constellations
- 48 Question
