Helmholtz Analytics Toolkit (Heat)
Heat is a distributed tensor framework for high performance data analytics. It is an offspring of the Helmholtz Analytics Framework (HAF) project conducted by several members of the Helmholtz Association of German Research Centres.
The goal of Heat is to fill the gap between machine learning libraries that have a strong focus on exploiting GPUs for performance, and traditional, distributed high-performance computing (HPC). The basic idea is to provide a dtype, distributed tensor library with machine learning methods based on it.
Features
- high-performance n-dimensional tensors
- CPU, GPU and distributed computation using MPI
- powerful machine learning methods using above mentioned tensors
Open Source
Heat is available as open source software. The source code is hosted on GitHub. Documentation is available on ReadTheDocs.
Last Modified: 12.10.2023