6. Next steps

• Full documentation for the Poplar software is available on the Graphcore documentation site: https://docs.graphcore.ai/. More information can be found on the Graphcore developer pages: https://www.graphcore.ai/developer.

• For an overview of the Poplar SDK and development tools, see the Poplar SDK Overview.

• The IPU Programmer’s Guide provides an introduction to the IPU architecture, programming model and tools available.

• The following quick start guides can help you get started with running an application:

  • PyTorch Quick Start

  • Jupyter Notebook Quick Start

  • Profiling Quick Start

  • TensorFlow 2 Quick Start

  • TensorFlow 1 Quick Start

  • PopART Quick Start

  • Poplar Quick Start

• Graphcore also has GitHub repositories with further examples:

  • https://github.com/graphcore/examples:

    • TensorFlow, PyTorch and PopART versions of commonly used machine learning models for training and inference, including CNNs such as ResNet, ResNeXt and EfficientNet

    • Training data for the above models

  • https://github.com/graphcore/tutorials:

    • Tutorials

    • Examples of using Poplar and IPU features

    • Examples of simple models

    • Source code from videos, blogs and other documents

    • Benchmarks for performance testing of layer types on your IPU system

• When looking for answers or asking questions on StackOverflow, use the tag “ipu”. For framework-specific questions, add the corresponding tag: “poplar”, “pytorch” and “popart”.

• Support is available from the Graphcore customer engineering team via the Graphcore support portal.

• For general help, discussions and announcements, please join our Graphcore Slack Community.