Extending Deep Learning Reference Stack Capabilities
Intel’s Mark Skarpness shares the next release of the integrated, highly-performant open sourced Deep Learning Reference Stack, Version 2.0 addresses key learnings and feedback and adds support for new tools, use cases, and workloads. As with the initial release, this version is highly-tuned and built for cloud native environments.
The update enables developers to shorten time between prototype and production by reducing complexity typical of deep learning software components. This developer friendly environment, along with deep learning and distributed training frameworks maintain flexibility to customize solutions.
Three key feature enhancements include PyTorch*, a machine learning library customers can use across deep learning applications; training framework Horovod*, for distributed deep learning; and the multi-use Jupyter Notebooks* which allows developers to use notebooks for a variety of simulations and modeling.
*Other names and brands may be claimed as the property of others. The nominative use of third party logos serves only the purposes of description and identification.