Logo
Optimising for the IPU: Computational Graph Recompilation and Executable Switching in TensorFlow
Version: 1.0.0
  • 1. Introduction
  • 2. Consideration 0: Avoiding recompilation
  • 3. Consideration 1: Sessions
  • 4. Consideration 2: Computational graphs
  • 5. Consideration 3: Batch size
  • 6. Consideration 4: Weights
  • 7. Consideration 5: Constants
  • 8. Consideration 6: Deep dive
  • 9. Code example
  • 10. Trademarks & copyright
Optimising for the IPU: Computational Graph Recompilation and Executable Switching in TensorFlow

10. Trademarks & copyright

Graphcore® and Poplar® are registered trademarks of Graphcore Ltd.

AI-Float™, Colossus™, Exchange Memory™, In-Processor-Memory™, IPU-Core™, IPU-Exchange™, IPU-Fabric™, IPU-Link™, IPU-M2000™, IPU-Machine™, IPU-POD™, IPU-Tile™, PopART™, PopLibs™, PopVision™, PopTorch™, Streaming Memory™ and Virtual-IPU™ are trademarks of Graphcore Ltd.

All other trademarks are the property of their respective owners.

Copyright © 2016-2020 Graphcore Ltd. All rights reserved.

Previous

Revision b1026b87.