The Holoscan SDK: building high-performance AI streaming applications


Keywords
AI, CUDA, medical, streaming, HPC, nvidia, cloud, edge, embedded, gpu, jetson, medical-devices, sensor
License
Apache-2.0
Install
pip install holoscan==2.4.0

Documentation

Holoscan SDK

The Holoscan SDK is part of NVIDIA Holoscan, the AI sensor processing platform that combines hardware systems for low-latency sensor and network connectivity, optimized libraries for data processing and AI, and core microservices to run streaming, imaging, and other applications, from embedded to edge to cloud. It can be used to build streaming AI pipelines for a variety of domains, including Medical Devices, High Performance Computing at the Edge, Industrial Inspection and more.

Table of Contents

Getting Started

Visit the Holoscan User Guide to get started with the Holoscan SDK: https://docs.nvidia.com/holoscan/sdk-user-guide/getting_started.html

The Holoscan User Guide includes:

  • An introduction to the NVIDIA Holoscan platform, including the Holoscan C++/Python SDK;
  • Requirements and setup steps;
  • Detailed SDK documentation, including a developer introduction, examples, and API details.

We also recommend visiting NVIDIA HoloHub to view community projects and reusable components available for your Holoscan project.

Obtaining the Holoscan SDK

The Holoscan User Guide documents several options to install and run the Holoscan SDK:

Visit the Holoscan User Guide for guidance to help choose which installation option may be right for your use case.

If the options above do not support your use case, you may prefer to build the SDK from source.

Please review Holoscan SDK prerequisites before getting started.

Troubleshooting and Feedback

We appreciate community discussion and feedback in support of Holoscan platform users and developers. We ask that users:

Contributing to Holoscan SDK

Holoscan SDK is developed internally and released as open source software. We welcome community contributions and may include them in Holoscan SDK releases at our discretion. Please refer to the Holoscan SDK Contributing Guide for more information.

Additional Notes

Relation to NVIDIA Clara

In previous releases, the prefix Clara was used to define Holoscan as a platform designed initially for medical devices. Starting with version 0.4.0, the Holoscan SDK is built to be domain-agnostic and can be used to build sensor AI applications in multiple domains. Domain specific content will be hosted on the HoloHub repository.

Repository structure

The repository is organized as such:

  • cmake/: CMake configuration files
  • data/: directory where data will be downloaded
  • examples/: source code for the examples
  • gxf_extensions/: source code for the holoscan SDK gxf codelets
  • include/: source code for the holoscan SDK core
  • modules/: source code for the holoscan SDK modules
  • patches/: patch files applied to dependencies
  • python/: python bindings for the holoscan SDK
  • scripts/: utility scripts
  • src/: source code for the holoscan SDK core
  • tests/: tests for the holoscan SDK