Software Documents

Licensed Software

This software is made available under the terms of the Graphcore End User License Agreement (EULA). Please ensure you have read and accept the terms of the license before using the software.

Release notes for each software release can be found on the Graphcore software download site.

Getting Started

Information on how to install the drivers and other essential software for your IPU hardware, how to run your first IPU program and get started with frameworks such as TensorFlow or PyTorch:

• Getting Started with IPU-POD Systems

• Getting Started with IPU-POD4 DA and IPU-POD16 DA

• Getting Started with Graphcloud Preview

PopRun and PopDist support running applications across multiple IPUs:

• PopRun and PopDist User Guide

An introduction to the IPU architecture, programming model and tools available:

• IPU Programmer’s Guide

A description of the components of the Poplar SDK and instructions for downloading and installing:

• SDK Overview

A set of pre-built Docker packages containing components of the Poplar SDK:

• Using IPUs from Docker

A dictionary of specialised terms related to Graphcore technology:

• Graphcore Glossary

TensorFlow

User guides and API reference for the IPU implementation of TensorFlow and Keras:

• TensorFlow 1 User Guide

• TensorFlow 2 User Guide

• Keras for the IPU

Technical Notes

A summary of the process of porting your TensorFlow application to the IPU:

• Porting TensorFlow models to the IPU

Exploiting the parallelism of the IPUs for your TensorFlow application:

• Model parallelism with TensorFlow: sharding and pipelining

Tips for managing graph compilation in TensorFlow:

• Optimising for the IPU: Computational Graph Recompilation and Executable Switching in TensorFlow

PyTorch

Support for PyTorch on the IPU:

• PyTorch for the IPU: User Guide

PopART

The Poplar Advanced Runtime (PopART) for importing and executing models from industry standard ML frameworks, using the ONNX format:

• PopART User Guide

• PopART Python API Reference

• PopART C++ API Reference

Poplar Graph Programming Framework

Information on how to use the Poplar graph programming tools to write code for the IPU:

• Poplar and PopLibs User Guide

Details of the functions in the Poplar and PopLibs libraries provided in the Poplar SDK:

• Poplar and PopLibs API Reference

Information about programming in vertices on the IPU. This also includes some useful detail about memory layout, programming in assembly language and interfaces to other tools:

• Vertex Programming Guide

Profiling and Debugging

The documentation for the PopVision Graph Analyser and System Analyser. This information is also available as context-sensitive help in the tools.

• PopVision User Guide

The contents of the files created by the Poplar tools with static and runtime profiling information. These files are mainly intended for use by the PopVision Graph Analyser but this may be useful to user’s wanting to understand the data for their own purposes or tool development:

• Poplar Profiling File Formats

Open Source Software

The following software is available as open source:

• TensorFlow for the IPU

  • Source repository

• PopTorch

  • Source repository

  • Build instructions

• PopART

  • Source repository

  • Build instructions

• PopLibs libraries

  • Source repository

  • Build instructions

• Poprithms: a library of graph algorithms used by the ML frameworks.

  • Source repository

  • Build instructions

See also the Examples and Tutorials.

License Agreements

• Graphcore End User License Agreement (EULA)

• PopLibs, PopTorch, PopART and Poprithms are licensed under the terms of the MIT license

• TensorFlow for the IPU is licensed under the Apache License 2.0

• Open source Contribution License Agreement (CLA)